2023-08-23 14:29:09 +00:00
|
|
|
#!/usr/bin/env python3
|
2023-08-21 20:07:43 +00:00
|
|
|
|
|
|
|
import gguf
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
import argparse
|
|
|
|
import concurrent.futures
|
|
|
|
import copy
|
|
|
|
import enum
|
|
|
|
import faulthandler
|
|
|
|
import functools
|
|
|
|
import io
|
|
|
|
import itertools
|
|
|
|
import json
|
|
|
|
import math
|
|
|
|
import mmap
|
|
|
|
import pickle
|
|
|
|
import re
|
|
|
|
import signal
|
|
|
|
import struct
|
|
|
|
import sys
|
|
|
|
import zipfile
|
2023-08-21 20:07:43 +00:00
|
|
|
import numpy as np
|
|
|
|
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
from abc import ABCMeta, abstractmethod
|
|
|
|
from dataclasses import dataclass
|
|
|
|
from pathlib import Path
|
2023-08-21 20:07:43 +00:00
|
|
|
from typing import (IO, TYPE_CHECKING, Any, Callable, Dict, Iterable, List, Literal, Optional, Sequence, Tuple, TypeVar, Union)
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
from sentencepiece import SentencePieceProcessor # type: ignore
|
2023-04-14 12:23:21 +00:00
|
|
|
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
if TYPE_CHECKING:
|
|
|
|
from typing_extensions import TypeAlias
|
|
|
|
|
|
|
|
if hasattr(faulthandler, 'register') and hasattr(signal, 'SIGUSR1'):
|
|
|
|
faulthandler.register(signal.SIGUSR1)
|
|
|
|
|
|
|
|
NDArray: 'TypeAlias' = 'np.ndarray[Any, Any]'
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
ARCH=gguf.MODEL_ARCH.LLAMA
|
|
|
|
NAMES=gguf.MODEL_TENSOR_NAMES[ARCH]
|
|
|
|
|
|
|
|
#
|
|
|
|
# data types
|
|
|
|
#
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
@dataclass(frozen=True)
|
|
|
|
class UnquantizedDataType:
|
|
|
|
name: str
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
DT_F16 = UnquantizedDataType('F16')
|
|
|
|
DT_F32 = UnquantizedDataType('F32')
|
|
|
|
DT_I32 = UnquantizedDataType('I32')
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
DT_BF16 = UnquantizedDataType('BF16')
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
DataType = Union[UnquantizedDataType]
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
DATA_TYPE_TO_NUMPY: Dict[DataType, 'np.dtype[Any]'] = {
|
2023-05-04 16:54:37 +00:00
|
|
|
DT_BF16: np.dtype(np.uint16),
|
2023-08-21 20:07:43 +00:00
|
|
|
DT_F16: np.dtype(np.float16),
|
|
|
|
DT_F32: np.dtype(np.float32),
|
|
|
|
DT_I32: np.dtype(np.int32),
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
NUMPY_TYPE_TO_DATA_TYPE: Dict['np.dtype[Any]', DataType] = \
|
|
|
|
{dtype: data_type for (data_type, dtype) in DATA_TYPE_TO_NUMPY.items()}
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
SAFETENSORS_DATA_TYPES: Dict[str, DataType] = {
|
|
|
|
'BF16': DT_BF16,
|
|
|
|
'F16': DT_F16,
|
|
|
|
'F32': DT_F32,
|
|
|
|
'I32': DT_I32,
|
|
|
|
}
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
2023-08-22 17:05:59 +00:00
|
|
|
# TODO: match this with `llama_ftype`
|
|
|
|
# TODO: rename to LLAMAFileType
|
|
|
|
# TODO: move to `gguf.py`
|
|
|
|
class GGMLFileType(enum.IntEnum):
|
2023-08-21 20:07:43 +00:00
|
|
|
AllF32 = 0
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
MostlyF16 = 1 # except 1d tensors
|
|
|
|
|
|
|
|
def type_for_tensor(self, name: str, tensor: 'LazyTensor') -> DataType:
|
|
|
|
if len(tensor.shape) == 1:
|
|
|
|
# 1D tensors are always F32.
|
|
|
|
return DT_F32
|
|
|
|
elif self == GGMLFileType.AllF32:
|
|
|
|
return DT_F32
|
|
|
|
elif self == GGMLFileType.MostlyF16:
|
|
|
|
return DT_F16
|
|
|
|
else:
|
|
|
|
raise ValueError(self)
|
|
|
|
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
#
|
|
|
|
# hparams loading
|
|
|
|
#
|
2023-06-22 12:20:47 +00:00
|
|
|
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
@dataclass
|
|
|
|
class Params:
|
2023-08-21 20:07:43 +00:00
|
|
|
n_vocab: int
|
|
|
|
n_embd: int
|
|
|
|
n_mult: int
|
|
|
|
n_layer: int
|
|
|
|
n_ctx: int
|
|
|
|
n_ff: int
|
|
|
|
n_head: int
|
|
|
|
n_head_kv: int
|
|
|
|
f_norm_eps: float
|
|
|
|
|
2023-08-24 18:04:05 +00:00
|
|
|
f_rope_freq_base: Optional[float] = None
|
|
|
|
|
2023-08-22 17:05:59 +00:00
|
|
|
ftype: Optional[GGMLFileType] = None
|
|
|
|
|
2023-08-24 16:26:19 +00:00
|
|
|
# path to the directory containing the model files
|
|
|
|
path_model: Optional['Path'] = None
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
@staticmethod
|
|
|
|
def find_n_mult(n_ff: int, n_embd: int) -> int:
|
|
|
|
# hardcoded magic range
|
|
|
|
for n_mult in range(8192, 1, -1):
|
|
|
|
calc_ff = (((8*n_embd) // 3 + n_mult - 1) // n_mult)*n_mult
|
|
|
|
if calc_ff == n_ff:
|
|
|
|
return n_mult
|
|
|
|
raise Exception(f"failed to find n_mult for (n_ff={n_ff}, n_embd={n_embd}).")
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
@staticmethod
|
2023-06-22 12:20:47 +00:00
|
|
|
def guessed(model: 'LazyModel') -> 'Params':
|
|
|
|
# try transformer naming first
|
|
|
|
n_vocab, n_embd = model["model.embed_tokens.weight"].shape if "model.embed_tokens.weight" in model else model["tok_embeddings.weight"].shape
|
|
|
|
|
|
|
|
# try transformer naming first
|
|
|
|
if "model.layers.0.self_attn.q_proj.weight" in model:
|
|
|
|
n_layer=next(i for i in itertools.count() if f"model.layers.{i}.self_attn.q_proj.weight" not in model)
|
2023-07-06 16:23:49 +00:00
|
|
|
elif "model.layers.0.self_attn.W_pack.weight" in model: # next: try baichuan naming
|
|
|
|
n_layer=next(i for i in itertools.count() if f"model.layers.{i}.self_attn.W_pack.weight" not in model)
|
2023-06-22 12:20:47 +00:00
|
|
|
else:
|
|
|
|
n_layer=next(i for i in itertools.count() if f"layers.{i}.attention.wq.weight" not in model)
|
|
|
|
|
2023-07-06 16:23:49 +00:00
|
|
|
if n_layer < 1:
|
|
|
|
raise Exception("failed to guess 'n_layer'. This model is unknown or unsupported.\n"
|
|
|
|
"Suggestion: provide 'config.json' of the model in the same directory containing model files.")
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
n_head = n_embd // 128 # guessed
|
|
|
|
n_mult = 256 # guessed
|
|
|
|
|
|
|
|
# TODO: verify this
|
|
|
|
n_ff = int(2 * (4 * n_embd) / 3)
|
|
|
|
n_ff = n_mult * ((n_ff + n_mult - 1) // n_mult)
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
return Params(
|
2023-08-21 20:07:43 +00:00
|
|
|
n_vocab = n_vocab,
|
|
|
|
n_embd = n_embd,
|
|
|
|
n_mult = n_mult,
|
|
|
|
n_layer = n_layer,
|
|
|
|
n_ctx = -1,
|
|
|
|
n_ff = n_ff,
|
|
|
|
n_head = n_head,
|
|
|
|
n_head_kv = n_head,
|
|
|
|
f_norm_eps = 1e-5,
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
)
|
|
|
|
|
2023-06-22 12:20:47 +00:00
|
|
|
@staticmethod
|
|
|
|
def loadHFTransformerJson(model: 'LazyModel', config_path: 'Path') -> 'Params':
|
|
|
|
config = json.load(open(config_path))
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
n_vocab = config["vocab_size"]
|
|
|
|
n_embd = config["hidden_size"]
|
|
|
|
n_layer = config["num_hidden_layers"]
|
|
|
|
n_ff = config["intermediate_size"]
|
|
|
|
n_head = config["num_attention_heads"]
|
|
|
|
n_head_kv = config["num_key_value_heads"] if "num_key_value_heads" in config else n_head
|
|
|
|
f_norm_eps = config["rms_norm_eps"]
|
2023-06-22 12:20:47 +00:00
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
n_mult = Params.find_n_mult(n_ff, n_embd)
|
|
|
|
|
|
|
|
if "max_sequence_length" in config:
|
|
|
|
n_ctx = config["max_sequence_length"]
|
|
|
|
elif "max_position_embeddings" in config:
|
|
|
|
n_ctx = config["max_position_embeddings"]
|
|
|
|
else:
|
|
|
|
raise Exception("failed to guess 'n_ctx'. This model is unknown or unsupported.\n"
|
|
|
|
"Suggestion: provide 'config.json' of the model in the same directory containing model files.")
|
2023-06-22 12:20:47 +00:00
|
|
|
|
|
|
|
return Params(
|
2023-08-21 20:07:43 +00:00
|
|
|
n_vocab = n_vocab,
|
|
|
|
n_embd = n_embd,
|
|
|
|
n_mult = n_mult,
|
|
|
|
n_layer = n_layer,
|
|
|
|
n_ctx = n_ctx,
|
|
|
|
n_ff = n_ff,
|
|
|
|
n_head = n_head,
|
|
|
|
n_head_kv = n_head_kv,
|
|
|
|
f_norm_eps = f_norm_eps,
|
2023-07-23 12:09:47 +00:00
|
|
|
)
|
|
|
|
|
|
|
|
# LLaMA v2 70B params.json
|
|
|
|
# {"dim": 8192, "multiple_of": 4096, "ffn_dim_multiplier": 1.3, "n_heads": 64, "n_kv_heads": 8, "n_layers": 80, "norm_eps": 1e-05, "vocab_size": -1
|
|
|
|
@staticmethod
|
|
|
|
def loadOriginalParamsJson(model: 'LazyModel', config_path: 'Path') -> 'Params':
|
|
|
|
config = json.load(open(config_path))
|
|
|
|
|
2023-08-24 18:04:05 +00:00
|
|
|
n_vocab = config["vocab_size"] if "vocab_size" in config else -1
|
|
|
|
n_embd = config["dim"]
|
|
|
|
n_layer = config["n_layers"]
|
|
|
|
n_mult = config["multiple_of"]
|
|
|
|
n_ff = -1
|
|
|
|
n_head = config["n_heads"]
|
|
|
|
n_head_kv = config["n_kv_heads"] if "n_kv_heads" in config else n_head
|
|
|
|
f_norm_eps = config["norm_eps"]
|
|
|
|
f_rope_freq_base = config["rope_theta"] if "rope_theta" in config else None
|
2023-07-23 12:09:47 +00:00
|
|
|
|
2023-08-24 19:10:39 +00:00
|
|
|
# hack to determine LLaMA v1 vs v2 vs CodeLlama
|
|
|
|
if f_rope_freq_base and f_rope_freq_base == 1000000:
|
|
|
|
# CodeLlama
|
|
|
|
n_ctx = 16384
|
|
|
|
elif config["norm_eps"] == 1e-05:
|
|
|
|
# LLaMA v2
|
|
|
|
n_ctx = 4096
|
|
|
|
else:
|
|
|
|
# LLaMA v1
|
|
|
|
n_ctx = 2048
|
|
|
|
|
2023-07-23 12:09:47 +00:00
|
|
|
if n_vocab == -1:
|
|
|
|
n_vocab = model["tok_embeddings.weight"].shape[0]
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
if n_ff == -1:
|
|
|
|
n_ff = model["layers.0.feed_forward.w1.weight"].shape[0]
|
|
|
|
|
2023-07-23 12:09:47 +00:00
|
|
|
return Params(
|
2023-08-24 18:04:05 +00:00
|
|
|
n_vocab = n_vocab,
|
|
|
|
n_embd = n_embd,
|
|
|
|
n_mult = n_mult,
|
|
|
|
n_layer = n_layer,
|
|
|
|
n_ctx = n_ctx,
|
|
|
|
n_ff = n_ff,
|
|
|
|
n_head = n_head,
|
|
|
|
n_head_kv = n_head_kv,
|
|
|
|
f_norm_eps = f_norm_eps,
|
|
|
|
f_rope_freq_base = f_rope_freq_base,
|
2023-06-22 12:20:47 +00:00
|
|
|
)
|
|
|
|
|
|
|
|
@staticmethod
|
|
|
|
def load(model_plus: 'ModelPlus') -> 'Params':
|
2023-07-23 12:09:47 +00:00
|
|
|
hf_config_path = model_plus.paths[0].parent / "config.json"
|
2023-06-22 12:20:47 +00:00
|
|
|
orig_config_path = model_plus.paths[0].parent / "params.json"
|
|
|
|
|
2023-07-23 12:09:47 +00:00
|
|
|
if hf_config_path.exists():
|
|
|
|
params = Params.loadHFTransformerJson(model_plus.model, hf_config_path)
|
|
|
|
elif orig_config_path.exists():
|
|
|
|
params = Params.loadOriginalParamsJson(model_plus.model, orig_config_path)
|
2023-06-22 12:20:47 +00:00
|
|
|
else:
|
|
|
|
params = Params.guessed(model_plus.model)
|
|
|
|
|
2023-08-24 16:26:19 +00:00
|
|
|
params.path_model = model_plus.paths[0].parent
|
|
|
|
|
2023-06-22 12:20:47 +00:00
|
|
|
return params
|
|
|
|
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
#
|
|
|
|
# vocab
|
|
|
|
#
|
|
|
|
|
|
|
|
class BpeVocab:
|
|
|
|
def __init__(self, fname_tokenizer: Path, fname_added_tokens: Optional[Path]) -> None:
|
|
|
|
self.bpe_tokenizer = json.loads(open(str(fname_tokenizer), encoding="utf-8").read())
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
added_tokens: Dict[str, int]
|
|
|
|
if fname_added_tokens is not None:
|
2023-08-21 20:07:43 +00:00
|
|
|
added_tokens = json.load(open(fname_added_tokens, encoding="utf-8"))
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
else:
|
|
|
|
added_tokens = {}
|
2023-08-21 20:07:43 +00:00
|
|
|
|
|
|
|
vocab_size: int = len(self.bpe_tokenizer)
|
|
|
|
expected_ids = list(range(vocab_size, vocab_size + len(added_tokens)))
|
|
|
|
actual_ids = sorted(added_tokens.values())
|
|
|
|
if expected_ids != actual_ids:
|
|
|
|
raise Exception(f"Expected added token IDs to be sequential and start at {len(added_tokens)}; got {actual_ids}")
|
|
|
|
|
|
|
|
items = sorted(added_tokens.items(), key=lambda text_idx: text_idx[1])
|
|
|
|
self.added_tokens_list = [text for (text, idx) in items]
|
|
|
|
self.vocab_size_base: int = vocab_size
|
|
|
|
self.vocab_size: int = self.vocab_size_base + len(self.added_tokens_list)
|
|
|
|
self.fname_tokenizer = fname_tokenizer
|
|
|
|
self.fname_added_tokens = fname_added_tokens
|
|
|
|
|
|
|
|
def bpe_tokens(self) -> Iterable[Tuple[bytes, float, gguf.TokenType]]:
|
|
|
|
tokenizer = self.bpe_tokenizer
|
|
|
|
from transformers.models.gpt2 import tokenization_gpt2
|
|
|
|
byte_encoder = tokenization_gpt2.bytes_to_unicode()
|
|
|
|
byte_decoder = {v: k for k, v in byte_encoder.items()}
|
|
|
|
for i, item in enumerate(tokenizer):
|
|
|
|
text: bytes = item.encode("utf-8")
|
|
|
|
score: float = -i
|
|
|
|
yield text, score, gguf.TokenType.USER_DEFINED
|
|
|
|
|
|
|
|
def added_tokens(self) -> Iterable[Tuple[bytes, float, gguf.TokenType]]:
|
|
|
|
for text in self.added_tokens_list:
|
|
|
|
score = -1000.0
|
|
|
|
yield text.encode("utf-8"), score, gguf.TokenType.USER_DEFINED
|
|
|
|
|
|
|
|
def all_tokens(self) -> Iterable[Tuple[bytes, float, gguf.TokenType]]:
|
|
|
|
yield from self.bpe_tokens()
|
|
|
|
yield from self.added_tokens()
|
|
|
|
|
|
|
|
def __repr__(self) -> str:
|
|
|
|
return f"BpeVocab with {self.vocab_size_base} base tokens and {len(self.added_tokens_list)} added tokens>"
|
|
|
|
|
|
|
|
|
|
|
|
class SentencePieceVocab:
|
|
|
|
def __init__(self, fname_tokenizer: Path, fname_added_tokens: Optional[Path]) -> None:
|
|
|
|
self.sentencepiece_tokenizer = SentencePieceProcessor(str(fname_tokenizer))
|
|
|
|
added_tokens: Dict[str, int]
|
|
|
|
if fname_added_tokens is not None:
|
|
|
|
added_tokens = json.load(open(fname_added_tokens, encoding="utf-8"))
|
2023-07-25 13:22:09 +00:00
|
|
|
else:
|
2023-08-21 20:07:43 +00:00
|
|
|
added_tokens = {}
|
|
|
|
|
|
|
|
vocab_size: int = self.sentencepiece_tokenizer.vocab_size()
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
expected_ids = list(range(vocab_size, vocab_size + len(added_tokens)))
|
2023-08-21 20:07:43 +00:00
|
|
|
actual_ids = sorted(added_tokens.values())
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
if expected_ids != actual_ids:
|
|
|
|
raise Exception(f"Expected added token IDs to be sequential and start at {len(added_tokens)}; got {actual_ids}")
|
2023-08-21 20:07:43 +00:00
|
|
|
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
items = sorted(added_tokens.items(), key=lambda text_idx: text_idx[1])
|
|
|
|
self.added_tokens_list = [text for (text, idx) in items]
|
|
|
|
self.vocab_size_base: int = vocab_size
|
|
|
|
self.vocab_size: int = self.vocab_size_base + len(self.added_tokens_list)
|
|
|
|
self.fname_tokenizer = fname_tokenizer
|
|
|
|
self.fname_added_tokens = fname_added_tokens
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
def sentencepiece_tokens(self) -> Iterable[Tuple[bytes, float, gguf.TokenType]]:
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
tokenizer = self.sentencepiece_tokenizer
|
2023-08-21 20:07:43 +00:00
|
|
|
for i in range(tokenizer.vocab_size()):
|
|
|
|
piece = tokenizer.id_to_piece(i)
|
|
|
|
text: bytes = piece.encode("utf-8")
|
|
|
|
score: float = tokenizer.get_score(i)
|
|
|
|
|
|
|
|
toktype = gguf.TokenType.NORMAL
|
|
|
|
if tokenizer.is_unknown(i):
|
|
|
|
toktype = gguf.TokenType.UNKNOWN
|
|
|
|
if tokenizer.is_control(i):
|
|
|
|
toktype = gguf.TokenType.CONTROL
|
|
|
|
|
|
|
|
# NOTE: I think added_tokens are user defined.
|
|
|
|
# ref: https://github.com/google/sentencepiece/blob/master/src/sentencepiece_model.proto
|
|
|
|
# if tokenizer.is_user_defined(i): toktype = gguf.TokenType.USER_DEFINED
|
|
|
|
|
|
|
|
if tokenizer.is_unused(i):
|
|
|
|
toktype = gguf.TokenType.UNUSED
|
|
|
|
if tokenizer.is_byte(i):
|
|
|
|
toktype = gguf.TokenType.BYTE
|
|
|
|
|
|
|
|
yield text, score, toktype
|
|
|
|
|
|
|
|
def added_tokens(self) -> Iterable[Tuple[bytes, float, gguf.TokenType]]:
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
for text in self.added_tokens_list:
|
|
|
|
score = -1000.0
|
2023-08-21 20:07:43 +00:00
|
|
|
yield text.encode("utf-8"), score, gguf.TokenType.USER_DEFINED
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
def all_tokens(self) -> Iterable[Tuple[bytes, float, gguf.TokenType]]:
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
yield from self.sentencepiece_tokens()
|
|
|
|
yield from self.added_tokens()
|
|
|
|
|
|
|
|
def __repr__(self) -> str:
|
|
|
|
return f"<SentencePieceVocab with {self.vocab_size_base} base tokens and {len(self.added_tokens_list)} added tokens>"
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
Vocab = Union[BpeVocab, SentencePieceVocab]
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
#
|
|
|
|
# data loading
|
|
|
|
# TODO: reuse (probably move to gguf.py?)
|
|
|
|
#
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
def permute(weights: NDArray, n_head: int, n_head_kv: int) -> NDArray:
|
|
|
|
#print( "permute debug " + str(weights.shape[0]) + " x " + str(weights.shape[1]) + " nhead " + str(n_head) + " nheadkv " + str(n_kv_head) )
|
|
|
|
if n_head_kv is not None and n_head != n_head_kv:
|
|
|
|
n_head //= n_head_kv
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
return (weights.reshape(n_head, 2, weights.shape[0] // n_head // 2, *weights.shape[1:])
|
2023-07-27 20:39:17 +00:00
|
|
|
.swapaxes(1, 2)
|
|
|
|
.reshape(weights.shape))
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
|
|
|
|
class Tensor(metaclass=ABCMeta):
|
|
|
|
data_type: DataType
|
|
|
|
|
|
|
|
@abstractmethod
|
|
|
|
def astype(self, data_type: DataType) -> 'Tensor': ...
|
|
|
|
@abstractmethod
|
2023-08-21 20:07:43 +00:00
|
|
|
def permute(self, n_head: int, n_head_kv: int) -> 'Tensor': ...
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
@abstractmethod
|
2023-07-01 17:00:25 +00:00
|
|
|
def permute_part(self, n_part: int, n_head: int) -> 'UnquantizedTensor': ...
|
|
|
|
@abstractmethod
|
|
|
|
def part(self, n_part: int) -> 'UnquantizedTensor': ...
|
|
|
|
@abstractmethod
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
def to_ggml(self) -> 'GGMLCompatibleTensor': ...
|
|
|
|
|
|
|
|
|
2023-05-04 16:54:37 +00:00
|
|
|
def bf16_to_fp32(bf16_arr: np.ndarray) -> np.ndarray:
|
|
|
|
assert bf16_arr.dtype == np.uint16, f"Input array should be of dtype uint16, but got {bf16_arr.dtype}"
|
|
|
|
fp32_arr = bf16_arr.astype(np.uint32) << 16
|
|
|
|
return fp32_arr.view(np.float32)
|
|
|
|
|
|
|
|
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
class UnquantizedTensor(Tensor):
|
|
|
|
def __init__(self, ndarray: NDArray) -> None:
|
|
|
|
assert isinstance(ndarray, np.ndarray)
|
|
|
|
self.ndarray = ndarray
|
|
|
|
self.data_type = NUMPY_TYPE_TO_DATA_TYPE[ndarray.dtype]
|
|
|
|
|
|
|
|
def astype(self, data_type: DataType) -> Tensor:
|
|
|
|
dtype = DATA_TYPE_TO_NUMPY[data_type]
|
2023-05-04 16:54:37 +00:00
|
|
|
if self.data_type == DT_BF16:
|
|
|
|
self.ndarray = bf16_to_fp32(self.ndarray)
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
return UnquantizedTensor(self.ndarray.astype(dtype))
|
|
|
|
|
|
|
|
def to_ggml(self) -> 'UnquantizedTensor':
|
|
|
|
return self
|
|
|
|
|
2023-07-01 17:00:25 +00:00
|
|
|
def permute_part(self, n_part: int, n_head: int) -> 'UnquantizedTensor':
|
|
|
|
r = self.ndarray.shape[0] // 3
|
|
|
|
return UnquantizedTensor(permute(self.ndarray[r * n_part : r * n_part + r, ...], n_head))
|
|
|
|
|
|
|
|
def part(self, n_part: int) -> 'UnquantizedTensor':
|
|
|
|
r = self.ndarray.shape[0] // 3
|
|
|
|
return UnquantizedTensor(self.ndarray[r * n_part : r * n_part + r, ...])
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
def permute(self, n_head: int, n_head_kv: int) -> 'UnquantizedTensor':
|
|
|
|
return UnquantizedTensor(permute(self.ndarray, n_head, n_head_kv))
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
|
|
|
|
def load_unquantized(lazy_tensor: 'LazyTensor', expected_dtype: Any = None, convert: bool = False) -> NDArray:
|
|
|
|
tensor = lazy_tensor.load()
|
|
|
|
assert isinstance(tensor, UnquantizedTensor)
|
|
|
|
|
|
|
|
# double-check:
|
|
|
|
actual_shape = list(tensor.ndarray.shape)
|
|
|
|
assert actual_shape == lazy_tensor.shape, (actual_shape, lazy_tensor.shape)
|
|
|
|
if expected_dtype is not None and expected_dtype != tensor.ndarray.dtype:
|
|
|
|
if convert:
|
|
|
|
tensor.ndarray = tensor.ndarray.astype(expected_dtype)
|
|
|
|
else:
|
|
|
|
raise ValueError(f'expected this tensor to have dtype {expected_dtype}, got {tensor.ndarray.dtype}')
|
|
|
|
|
|
|
|
return tensor.ndarray
|
|
|
|
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
GGMLCompatibleTensor = Union[UnquantizedTensor]
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
|
|
|
|
class DeferredPermutedTensor(Tensor):
|
2023-08-21 20:07:43 +00:00
|
|
|
def __init__(self, base: Tensor, n_head: int, n_head_kv: int) -> None:
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
self.base = base
|
|
|
|
self.n_head = n_head
|
|
|
|
self.data_type = self.base.data_type
|
|
|
|
|
|
|
|
def astype(self, data_type: DataType) -> Tensor:
|
2023-08-21 20:07:43 +00:00
|
|
|
return self.base.astype(data_type).permute(self.n_head, self.n_head_kv)
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
def to_ggml(self) -> GGMLCompatibleTensor:
|
2023-08-21 20:07:43 +00:00
|
|
|
return self.base.to_ggml().permute(self.n_head, self.n_head_kv)
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
def permute(self, n_head: int, n_head_kv: int) -> Tensor:
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
raise Exception("shouldn't permute twice")
|
|
|
|
|
|
|
|
|
|
|
|
@dataclass
|
|
|
|
class LazyTensor:
|
|
|
|
_load: Callable[[], Tensor]
|
|
|
|
shape: List[int]
|
|
|
|
data_type: DataType
|
|
|
|
description: str
|
|
|
|
|
|
|
|
def load(self) -> Tensor:
|
|
|
|
ret = self._load()
|
|
|
|
assert ret.data_type == self.data_type, (self.data_type, ret.data_type, self.description)
|
|
|
|
return ret
|
|
|
|
|
|
|
|
def astype(self, data_type: DataType) -> 'LazyTensor':
|
|
|
|
self.validate_conversion_to(data_type)
|
|
|
|
|
|
|
|
def load() -> Tensor:
|
|
|
|
return self.load().astype(data_type)
|
|
|
|
return LazyTensor(load, self.shape, data_type, f'convert({data_type}) {self.description}')
|
|
|
|
|
|
|
|
def validate_conversion_to(self, data_type: DataType) -> None:
|
|
|
|
if data_type == self.data_type:
|
|
|
|
return
|
|
|
|
|
|
|
|
|
|
|
|
LazyModel = Dict[str, LazyTensor]
|
|
|
|
|
|
|
|
|
|
|
|
@dataclass
|
|
|
|
class ModelPlus:
|
|
|
|
model: LazyModel
|
|
|
|
paths: List[Path] # Where this was read from.
|
|
|
|
format: Literal['ggml', 'torch', 'safetensors']
|
|
|
|
vocab: Optional[Vocab] # For GGML models (which have vocab built in), the vocab.
|
|
|
|
|
|
|
|
|
|
|
|
def merge_sharded(models: List[LazyModel]) -> LazyModel:
|
|
|
|
# Original LLaMA models have each file contain one part of each tensor.
|
|
|
|
# Use a dict instead of a set to preserve order.
|
|
|
|
names = {name: None for model in models for name in model}
|
|
|
|
|
|
|
|
def convert(name: str) -> LazyTensor:
|
|
|
|
lazy_tensors: List[LazyTensor] = [model[name] for model in models]
|
|
|
|
if len(lazy_tensors) == 1:
|
|
|
|
# only one file; don't go through this procedure since there might
|
|
|
|
# be quantized tensors
|
|
|
|
return lazy_tensors[0]
|
|
|
|
if len(lazy_tensors[0].shape) == 1:
|
|
|
|
# the tensor is just duplicated in every file
|
|
|
|
return lazy_tensors[0]
|
|
|
|
if name.startswith('tok_embeddings.') or \
|
|
|
|
name.endswith('.attention.wo.weight') or \
|
|
|
|
name.endswith('.feed_forward.w2.weight'):
|
|
|
|
# split by columns
|
|
|
|
axis = 1
|
|
|
|
else:
|
|
|
|
# split by rows
|
|
|
|
axis = 0
|
|
|
|
concatenated_shape = list(lazy_tensors[0].shape)
|
|
|
|
concatenated_shape[axis] = sum(tensor.shape[axis] for tensor in lazy_tensors)
|
|
|
|
|
|
|
|
def load() -> UnquantizedTensor:
|
|
|
|
ndarrays = [load_unquantized(tensor) for tensor in lazy_tensors]
|
|
|
|
concatenated: NDArray = np.concatenate(ndarrays, axis=axis)
|
|
|
|
return UnquantizedTensor(concatenated)
|
|
|
|
description = 'concatenated[[' + '] | ['.join(lt.description for lt in lazy_tensors) + ']]'
|
|
|
|
return LazyTensor(load, concatenated_shape, lazy_tensors[0].data_type, description)
|
|
|
|
return {name: convert(name) for name in names}
|
|
|
|
|
|
|
|
|
|
|
|
def merge_multifile_models(models_plus: List[ModelPlus]) -> ModelPlus:
|
|
|
|
formats = set(mp.format for mp in models_plus)
|
|
|
|
assert len(formats) == 1, "different formats?"
|
|
|
|
format = formats.pop()
|
|
|
|
paths = [path for mp in models_plus for path in mp.paths]
|
|
|
|
# Use the first non-None vocab, if any.
|
|
|
|
try:
|
|
|
|
vocab = next(mp.vocab for mp in models_plus if mp.vocab is not None)
|
|
|
|
except StopIteration:
|
|
|
|
vocab = None
|
|
|
|
|
|
|
|
if any("model.embed_tokens.weight" in mp.model for mp in models_plus):
|
|
|
|
# Transformers models put different tensors in different files, but
|
|
|
|
# don't split indivdual tensors between files.
|
|
|
|
model: LazyModel = {}
|
|
|
|
for mp in models_plus:
|
|
|
|
model.update(mp.model)
|
|
|
|
else:
|
|
|
|
model = merge_sharded([mp.model for mp in models_plus])
|
|
|
|
|
|
|
|
return ModelPlus(model, paths, format, vocab)
|
|
|
|
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
def permute_lazy(lazy_tensor: LazyTensor, n_head: int, n_head_kv: int) -> LazyTensor:
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
def load() -> Tensor:
|
2023-08-21 20:07:43 +00:00
|
|
|
return lazy_tensor.load().permute(n_head, n_head_kv)
|
|
|
|
return LazyTensor(load, lazy_tensor.shape, lazy_tensor.data_type, f'permute({n_head}, {n_head_kv}) ' + lazy_tensor.description)
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
2023-07-01 17:00:25 +00:00
|
|
|
def permute_part_lazy(lazy_tensor: LazyTensor, n_part: int, n_head: int) -> LazyTensor:
|
|
|
|
def load() -> Tensor:
|
|
|
|
return lazy_tensor.load().permute_part(n_part, n_head)
|
|
|
|
s = lazy_tensor.shape.copy()
|
|
|
|
s[0] = s[0] // 3
|
|
|
|
return LazyTensor(load, s, lazy_tensor.data_type, f'permute({n_head}) ' + lazy_tensor.description)
|
|
|
|
|
|
|
|
def part_lazy(lazy_tensor: LazyTensor, n_part: int) -> LazyTensor:
|
|
|
|
def load() -> Tensor:
|
|
|
|
return lazy_tensor.load().part(n_part)
|
|
|
|
s = lazy_tensor.shape.copy()
|
|
|
|
s[0] = s[0] // 3
|
|
|
|
return LazyTensor(load, s, lazy_tensor.data_type, 'part ' + lazy_tensor.description)
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
|
|
|
|
# Functionality that simulates `torch.load` but where individual tensors are
|
|
|
|
# only loaded into memory on demand, not all at once.
|
|
|
|
# PyTorch can't do this natively as of time of writing:
|
|
|
|
# - https://github.com/pytorch/pytorch/issues/64327
|
|
|
|
# This allows us to de-shard without multiplying RAM usage, and also
|
|
|
|
# conveniently drops the PyTorch dependency (though we still need numpy).
|
|
|
|
|
|
|
|
|
|
|
|
@dataclass
|
|
|
|
class LazyStorageKind:
|
|
|
|
data_type: DataType
|
|
|
|
|
|
|
|
|
|
|
|
@dataclass
|
|
|
|
class LazyStorage:
|
|
|
|
load: Callable[[int, int], NDArray]
|
|
|
|
kind: LazyStorageKind
|
|
|
|
description: str
|
|
|
|
|
|
|
|
|
|
|
|
class LazyUnpickler(pickle.Unpickler):
|
|
|
|
def __init__(self, fp: IO[bytes], data_base_path: str, zip_file: zipfile.ZipFile):
|
|
|
|
super().__init__(fp)
|
|
|
|
self.data_base_path = data_base_path
|
|
|
|
self.zip_file = zip_file
|
|
|
|
|
|
|
|
def persistent_load(self, pid: Any) -> Any:
|
|
|
|
assert pid[0] == 'storage'
|
|
|
|
assert isinstance(pid[1], LazyStorageKind)
|
|
|
|
data_type = pid[1].data_type
|
|
|
|
filename_stem = pid[2]
|
|
|
|
filename = self.data_base_path + '/' + filename_stem
|
|
|
|
info = self.zip_file.getinfo(filename)
|
|
|
|
|
|
|
|
def load(offset: int, elm_count: int) -> NDArray:
|
|
|
|
dtype = DATA_TYPE_TO_NUMPY.get(data_type)
|
|
|
|
if dtype is None:
|
|
|
|
raise Exception("tensor stored in unsupported format")
|
|
|
|
fp = self.zip_file.open(info)
|
|
|
|
fp.seek(offset * dtype.itemsize)
|
|
|
|
size = elm_count * dtype.itemsize
|
|
|
|
data = fp.read(size)
|
|
|
|
assert len(data) == size
|
|
|
|
return np.frombuffer(data, dtype)
|
|
|
|
description = f'storage data_type={data_type} path-in-zip={filename} path={self.zip_file.filename}'
|
|
|
|
return LazyStorage(load=load, kind=pid[1], description=description)
|
|
|
|
|
2023-06-17 10:32:48 +00:00
|
|
|
# @staticmethod
|
|
|
|
def lazy_rebuild_tensor_v2(storage: Any, storage_offset: Any, size: Any, stride: Any,
|
|
|
|
# pyright: ignore[reportSelfClsParameterName]
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
requires_grad: Any, backward_hooks: Any, metadata: Any = None) -> LazyTensor:
|
|
|
|
assert isinstance(storage, LazyStorage)
|
|
|
|
|
|
|
|
def load() -> UnquantizedTensor:
|
|
|
|
elm_count = stride[0] * size[0]
|
|
|
|
return UnquantizedTensor(storage.load(storage_offset, elm_count).reshape(size))
|
|
|
|
description = f'pickled storage_offset={storage_offset} in {storage.description}'
|
|
|
|
return LazyTensor(load, list(size), storage.kind.data_type, description)
|
|
|
|
|
2023-05-05 00:17:07 +00:00
|
|
|
# @staticmethod
|
2023-05-04 16:54:37 +00:00
|
|
|
def rebuild_from_type_v2(func, new_type, args, state):
|
|
|
|
return func(*args)
|
|
|
|
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
CLASSES: Dict[Any, Any] = {
|
2023-05-04 16:54:37 +00:00
|
|
|
('torch._tensor', '_rebuild_from_type_v2'): rebuild_from_type_v2,
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
('torch._utils', '_rebuild_tensor_v2'): lazy_rebuild_tensor_v2,
|
|
|
|
('torch', 'BFloat16Storage'): LazyStorageKind(DT_BF16),
|
|
|
|
('torch', 'HalfStorage'): LazyStorageKind(DT_F16),
|
|
|
|
('torch', 'FloatStorage'): LazyStorageKind(DT_F32),
|
|
|
|
('torch', 'IntStorage'): LazyStorageKind(DT_I32),
|
2023-05-04 16:54:37 +00:00
|
|
|
('torch', 'Tensor'): LazyTensor,
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
def find_class(self, module: str, name: str) -> Any:
|
|
|
|
if not module.startswith('torch'):
|
|
|
|
return super().find_class(module, name)
|
|
|
|
return self.CLASSES[(module, name)]
|
|
|
|
|
|
|
|
|
|
|
|
def lazy_load_torch_file(outer_fp: IO[bytes], path: Path) -> ModelPlus:
|
|
|
|
zf = zipfile.ZipFile(outer_fp)
|
|
|
|
pickle_paths = [name for name in zf.namelist() if name.endswith('.pkl')]
|
|
|
|
assert len(pickle_paths) == 1, pickle_paths
|
|
|
|
pickle_fp = zf.open(pickle_paths[0], 'r')
|
|
|
|
unpickler = LazyUnpickler(pickle_fp,
|
|
|
|
data_base_path=pickle_paths[0][:-4],
|
|
|
|
zip_file=zf)
|
|
|
|
model = unpickler.load()
|
|
|
|
as_dict = dict(model.items())
|
|
|
|
return ModelPlus(model=as_dict, paths=[path], format='torch', vocab=None)
|
|
|
|
|
|
|
|
|
|
|
|
def lazy_load_safetensors_file(fp: IO[bytes], path: Path) -> ModelPlus:
|
|
|
|
header_size, = struct.unpack('<Q', fp.read(8))
|
|
|
|
header: Dict[str, Dict[str, Any]] = json.loads(fp.read(header_size))
|
|
|
|
# Use mmap for the actual data to avoid race conditions with the file offset.
|
|
|
|
mapped = memoryview(mmap.mmap(fp.fileno(), 0, access=mmap.ACCESS_READ))
|
2023-04-15 21:53:21 +00:00
|
|
|
byte_buf = mapped[8 + header_size:]
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
def convert(info: Dict[str, Any]) -> LazyTensor:
|
|
|
|
data_type = SAFETENSORS_DATA_TYPES[info['dtype']]
|
|
|
|
numpy_dtype = DATA_TYPE_TO_NUMPY[data_type]
|
|
|
|
shape: List[int] = info['shape']
|
|
|
|
begin, end = info['data_offsets']
|
|
|
|
assert 0 <= begin <= end <= len(byte_buf)
|
|
|
|
assert end - begin == math.prod(shape) * numpy_dtype.itemsize
|
|
|
|
buf = byte_buf[begin:end]
|
|
|
|
|
|
|
|
def load() -> UnquantizedTensor:
|
|
|
|
return UnquantizedTensor(np.frombuffer(buf, dtype=numpy_dtype).reshape(shape))
|
|
|
|
description = f'safetensors begin={begin} end={end} type={data_type} path={path}'
|
|
|
|
return LazyTensor(load, shape, data_type, description)
|
2023-05-08 11:54:26 +00:00
|
|
|
model = {name: convert(info) for (name, info) in header.items() if name != '__metadata__'}
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
return ModelPlus(model=model, paths=[path], format='safetensors', vocab=None)
|
|
|
|
|
|
|
|
|
|
|
|
def must_read(fp: IO[bytes], length: int) -> bytes:
|
|
|
|
ret = fp.read(length)
|
|
|
|
if len(ret) < length:
|
|
|
|
raise Exception("unexpectedly reached end of file")
|
|
|
|
return ret
|
|
|
|
|
|
|
|
|
|
|
|
@functools.lru_cache(maxsize=None)
|
|
|
|
def lazy_load_file(path: Path) -> ModelPlus:
|
|
|
|
fp = open(path, 'rb')
|
|
|
|
first8 = fp.read(8)
|
|
|
|
fp.seek(0)
|
|
|
|
if first8[:2] == b'PK':
|
|
|
|
# A zip file, i.e. PyTorch format
|
|
|
|
return lazy_load_torch_file(fp, path)
|
|
|
|
elif struct.unpack('<Q', first8)[0] < 16 * 1024 * 1024:
|
|
|
|
# Probably safetensors
|
|
|
|
return lazy_load_safetensors_file(fp, path)
|
|
|
|
else:
|
|
|
|
raise ValueError(f"unknown format: {path}")
|
|
|
|
|
|
|
|
|
|
|
|
In = TypeVar('In')
|
|
|
|
Out = TypeVar('Out')
|
|
|
|
|
|
|
|
def bounded_parallel_map(func: Callable[[In], Out], iterable: Iterable[In], concurrency: int) -> Iterable[Out]:
|
|
|
|
'''Parallel map, but with backpressure. If the caller doesn't call `next`
|
|
|
|
fast enough, this will stop calling `func` at some point rather than
|
|
|
|
letting results pile up in memory. Specifically, there is a max of one
|
|
|
|
output value buffered per thread.'''
|
|
|
|
with concurrent.futures.ThreadPoolExecutor() as executor:
|
|
|
|
futures: List[concurrent.futures.Future[Out]] = []
|
|
|
|
items_rev = list(iterable)[::-1]
|
|
|
|
for i in range(min(concurrency, len(items_rev))):
|
|
|
|
futures.append(executor.submit(func, items_rev.pop()))
|
|
|
|
while futures:
|
|
|
|
result = futures.pop(0).result()
|
|
|
|
if items_rev:
|
|
|
|
futures.append(executor.submit(func, items_rev.pop()))
|
|
|
|
yield result
|
|
|
|
|
|
|
|
|
|
|
|
def check_vocab_size(params: Params, vocab: Vocab) -> None:
|
|
|
|
if params.n_vocab != vocab.vocab_size:
|
2023-08-21 20:07:43 +00:00
|
|
|
assert isinstance(vocab, BpeVocab) or isinstance(vocab, SentencePieceVocab)
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
if params.n_vocab == vocab.vocab_size_base:
|
|
|
|
print("Ignoring added_tokens.json since model matches vocab size without it.")
|
|
|
|
vocab.added_tokens_list = []
|
|
|
|
vocab.vocab_size = vocab.vocab_size_base
|
|
|
|
return
|
|
|
|
msg = f"Vocab size mismatch (model has {params.n_vocab}, but {vocab.fname_tokenizer}"
|
|
|
|
if vocab.fname_added_tokens is not None:
|
|
|
|
msg += f" combined with {vocab.fname_added_tokens}"
|
|
|
|
msg += f" has {vocab.vocab_size})."
|
|
|
|
if vocab.vocab_size < params.n_vocab < vocab.vocab_size + 20 and vocab.fname_added_tokens is None:
|
|
|
|
msg += f" Most likely you are missing added_tokens.json (should be in {vocab.fname_tokenizer.parent})."
|
|
|
|
raise Exception(msg)
|
|
|
|
|
|
|
|
|
|
|
|
class OutputFile:
|
|
|
|
def __init__(self, fname_out: Path) -> None:
|
2023-08-21 20:07:43 +00:00
|
|
|
self.gguf = gguf.GGUFWriter(fname_out, gguf.MODEL_ARCH_NAMES[ARCH])
|
|
|
|
|
|
|
|
def add_meta_arch(self, params: Params) -> None:
|
2023-08-24 16:26:19 +00:00
|
|
|
name = "LLaMA"
|
2023-08-23 20:08:04 +00:00
|
|
|
if (params.n_ctx == 4096):
|
2023-08-24 16:26:19 +00:00
|
|
|
name = "LLaMA v2"
|
|
|
|
if params.path_model:
|
|
|
|
name = str(params.path_model.parent).split('/')[-1]
|
2023-08-23 20:08:04 +00:00
|
|
|
|
2023-08-24 16:26:19 +00:00
|
|
|
self.gguf.add_name (name)
|
2023-08-21 20:07:43 +00:00
|
|
|
self.gguf.add_context_length (params.n_ctx)
|
|
|
|
self.gguf.add_embedding_length (params.n_embd)
|
|
|
|
self.gguf.add_block_count (params.n_layer)
|
|
|
|
self.gguf.add_feed_forward_length (params.n_ff)
|
|
|
|
self.gguf.add_rope_dimension_count(params.n_embd // params.n_head)
|
|
|
|
self.gguf.add_head_count (params.n_head)
|
|
|
|
self.gguf.add_head_count_kv (params.n_head_kv)
|
|
|
|
self.gguf.add_layer_norm_rms_eps (params.f_norm_eps)
|
|
|
|
|
2023-08-24 18:04:05 +00:00
|
|
|
if params.f_rope_freq_base:
|
|
|
|
self.gguf.add_rope_freq_base(params.f_rope_freq_base)
|
|
|
|
|
2023-08-22 17:05:59 +00:00
|
|
|
if params.ftype:
|
|
|
|
self.gguf.add_file_type(params.ftype)
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
def add_meta_vocab(self, vocab: Vocab) -> None:
|
|
|
|
tokens = []
|
|
|
|
scores = []
|
|
|
|
toktypes = []
|
|
|
|
# NOTE: `all_tokens` returns the the base vocabulary and added tokens
|
|
|
|
# TODO: add special tokens?
|
|
|
|
for text, score, toktype in vocab.all_tokens():
|
|
|
|
tokens.append(text)
|
|
|
|
scores.append(score)
|
|
|
|
toktypes.append(toktype)
|
|
|
|
|
|
|
|
self.gguf.add_tokenizer_model("llama")
|
|
|
|
self.gguf.add_token_list(tokens)
|
|
|
|
self.gguf.add_token_scores(scores)
|
|
|
|
self.gguf.add_token_types(toktypes)
|
|
|
|
|
|
|
|
def add_tensor_info(self, name: str, tensor: LazyTensor) -> None:
|
|
|
|
n_elements = 1
|
|
|
|
for dim in tensor.shape:
|
|
|
|
n_elements *= dim
|
|
|
|
data_type = DATA_TYPE_TO_NUMPY[tensor.data_type]
|
|
|
|
data_nbytes = n_elements * data_type.itemsize
|
|
|
|
self.gguf.add_tensor_info(name, tensor.shape, data_type, data_nbytes)
|
|
|
|
|
|
|
|
def write_meta(self) -> None:
|
|
|
|
self.gguf.write_header_to_file()
|
|
|
|
self.gguf.write_kv_data_to_file()
|
|
|
|
|
|
|
|
def write_tensor_info(self) -> None:
|
|
|
|
self.gguf.write_ti_data_to_file()
|
|
|
|
|
|
|
|
def close(self) -> None:
|
|
|
|
self.gguf.close()
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
@staticmethod
|
2023-08-21 20:07:43 +00:00
|
|
|
def write_vocab_only(fname_out: Path, params: Params, vocab: Vocab) -> None:
|
|
|
|
check_vocab_size(params, vocab)
|
|
|
|
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
of = OutputFile(fname_out)
|
2023-08-21 20:07:43 +00:00
|
|
|
|
|
|
|
# meta data
|
|
|
|
of.add_meta_arch(params)
|
|
|
|
of.add_meta_vocab(vocab)
|
|
|
|
of.write_meta()
|
|
|
|
|
|
|
|
of.close()
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
@staticmethod
|
2023-08-21 20:07:43 +00:00
|
|
|
def write_all(fname_out: Path, params: Params, model: LazyModel, vocab: Vocab) -> None:
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
check_vocab_size(params, vocab)
|
2023-08-21 20:07:43 +00:00
|
|
|
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
of = OutputFile(fname_out)
|
2023-08-21 20:07:43 +00:00
|
|
|
|
|
|
|
# meta data
|
|
|
|
of.add_meta_arch(params)
|
|
|
|
of.add_meta_vocab(vocab)
|
|
|
|
|
|
|
|
# tensor info
|
|
|
|
for name, lazy_tensor in model.items():
|
|
|
|
of.add_tensor_info(name, lazy_tensor)
|
|
|
|
|
|
|
|
of.write_meta()
|
|
|
|
of.write_tensor_info()
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
def do_item(item: Tuple[str, LazyTensor]) -> NDArray:
|
|
|
|
name, lazy_tensor = item
|
|
|
|
return lazy_tensor.load().to_ggml().ndarray
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
# tensor data
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
ndarrays = bounded_parallel_map(do_item, model.items(), concurrency=8)
|
|
|
|
for i, ((name, lazy_tensor), ndarray) in enumerate(zip(model.items(), ndarrays)):
|
2023-04-16 10:58:48 +00:00
|
|
|
size = ' x '.join(f"{dim:6d}" for dim in lazy_tensor.shape)
|
|
|
|
padi = len(str(len(model)))
|
|
|
|
print(f"[{i+1:{padi}d}/{len(model)}] Writing tensor {name:38s} | size {size:16} | type {lazy_tensor.data_type}")
|
2023-08-21 20:07:43 +00:00
|
|
|
of.gguf.write_tensor_data(ndarray)
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
of.close()
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
def pick_output_type(model: LazyModel, output_type_str: Optional[str]) -> GGMLFileType:
|
2023-08-21 20:07:43 +00:00
|
|
|
wq_type = model[NAMES[gguf.MODEL_TENSOR.ATTN_Q].format(bid=0)+".weight"].data_type
|
|
|
|
|
|
|
|
if output_type_str == "f32" or (output_type_str is None and wq_type == DT_F32):
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
return GGMLFileType.AllF32
|
2023-08-21 20:07:43 +00:00
|
|
|
if output_type_str == "f16" or (output_type_str is None and wq_type in (DT_F16, DT_BF16)):
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
return GGMLFileType.MostlyF16
|
2023-08-21 20:07:43 +00:00
|
|
|
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
name_to_type = {name: lazy_tensor.data_type for (name, lazy_tensor) in model.items()}
|
2023-08-21 20:07:43 +00:00
|
|
|
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
raise Exception(f"Unexpected combination of types: {name_to_type}")
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
def convert_to_output_type(model: LazyModel, output_type: GGMLFileType) -> LazyModel:
|
|
|
|
return {name: tensor.astype(output_type.type_for_tensor(name, tensor))
|
|
|
|
for (name, tensor) in model.items()}
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
def convert_model_names(model: LazyModel, params: Params) -> LazyModel:
|
|
|
|
tmap = gguf.get_tensor_name_map(ARCH, params.n_layer)
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
tmp = model
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
# HF models permut or pack some of the tensors, so we need to undo that
|
|
|
|
for i in itertools.count():
|
|
|
|
if f"model.layers.{i}.self_attn.q_proj.weight" in model:
|
|
|
|
print(f"Permuting layer {i}")
|
|
|
|
tmp[f"model.layers.{i}.self_attn.q_proj.weight"] = permute_lazy(model[f"model.layers.{i}.self_attn.q_proj.weight"], params.n_head, params.n_head)
|
|
|
|
tmp[f"model.layers.{i}.self_attn.k_proj.weight"] = permute_lazy(model[f"model.layers.{i}.self_attn.k_proj.weight"], params.n_head, params.n_head_kv)
|
|
|
|
#tmp[f"model.layers.{i}.self_attn.v_proj.weight"] = model[f"model.layers.{i}.self_attn.v_proj.weight"]
|
|
|
|
elif f"model.layers.{i}.self_attn.W_pack.weight" in model:
|
|
|
|
print(f"Unpacking and permuting layer {i}")
|
|
|
|
tmp[f"model.layers.{i}.self_attn.q_proj.weight"] = permute_part_lazy(model[f"model.layers.{i}.self_attn.W_pack.weight"], 0, params.n_head, params.n_head)
|
|
|
|
tmp[f"model.layers.{i}.self_attn.k_proj.weight"] = permute_part_lazy(model[f"model.layers.{i}.self_attn.W_pack.weight"], 1, params.n_head, params.n_head_kv)
|
|
|
|
tmp[f"model.layers.{i}.self_attn.v_proj.weight"] = part_lazy (model[f"model.layers.{i}.self_attn.W_pack.weight"], 2)
|
|
|
|
else:
|
|
|
|
break
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
out: LazyModel = {}
|
|
|
|
for name, lazy_tensor in model.items():
|
|
|
|
name_new = name
|
|
|
|
|
|
|
|
if name in tmap:
|
|
|
|
name_new = tmap[name]
|
|
|
|
elif name.endswith(".weight") and name[:-7] in tmap:
|
|
|
|
name_new = tmap[name[:-7]] + ".weight"
|
|
|
|
elif name.endswith(".bias") and name[:-5] in tmap:
|
|
|
|
name_new = tmap[name[:-5]] + ".bias"
|
|
|
|
else:
|
|
|
|
raise Exception(f"Unexpected tensor name: {name}")
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
if gguf.should_skip_tensor_TMP(ARCH, params.n_layer, name_new):
|
|
|
|
print(f"skipping tensor {name_new}")
|
|
|
|
continue
|
|
|
|
else:
|
|
|
|
print(f"{name:48s} -> {name_new:40s} | {lazy_tensor.data_type} | {lazy_tensor.shape}")
|
|
|
|
out[name_new] = lazy_tensor
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
return out
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
def nth_multifile_path(path: Path, n: int) -> Optional[Path]:
|
|
|
|
'''Given any path belonging to a multi-file model (e.g. foo.bin.1), return
|
|
|
|
the nth path in the model.
|
|
|
|
'''
|
|
|
|
# Support the following patterns:
|
|
|
|
patterns: List[Tuple[str, str]] = [
|
|
|
|
# - x.00.pth, x.01.pth, etc.
|
|
|
|
(r'\.[0-9]{2}\.pth$', f'.{n:02}.pth'),
|
|
|
|
# - x-00001-of-00002.bin, x-00002-of-00002.bin, etc.
|
|
|
|
(r'-[0-9]{5}-of-(.*)$', fr'-{n:05}-of-\1'),
|
|
|
|
# x.bin, x.bin.1, etc.
|
|
|
|
(r'(\.[0-9]+)?$', r'\1' if n == 0 else fr'\1.{n}')
|
|
|
|
]
|
|
|
|
for regex, replacement in patterns:
|
|
|
|
if re.search(regex, path.name):
|
|
|
|
new_path = path.with_name(re.sub(regex, replacement, path.name))
|
|
|
|
if new_path.exists():
|
|
|
|
return new_path
|
|
|
|
return None
|
|
|
|
|
|
|
|
|
|
|
|
def find_multifile_paths(path: Path) -> List[Path]:
|
|
|
|
'''Given any path belonging to a multi-file model (e.g. foo.bin.1), return
|
|
|
|
the whole list of paths in the model.
|
|
|
|
'''
|
|
|
|
ret: List[Path] = []
|
|
|
|
for i in itertools.count():
|
|
|
|
nth_path = nth_multifile_path(path, i)
|
|
|
|
if nth_path is None:
|
|
|
|
break
|
|
|
|
ret.append(nth_path)
|
|
|
|
if not ret:
|
|
|
|
# No matches. This should only happen if the file was named, e.g.,
|
|
|
|
# foo.0, and there was no file named foo. Oh well, try to process it
|
|
|
|
# as a single file.
|
|
|
|
return [path]
|
|
|
|
return ret
|
|
|
|
|
|
|
|
|
|
|
|
def load_some_model(path: Path) -> ModelPlus:
|
|
|
|
'''Load a model of any supported format.'''
|
|
|
|
# Be extra-friendly and accept either a file or a directory:
|
|
|
|
if path.is_dir():
|
2023-05-08 11:54:26 +00:00
|
|
|
# Check if it's a set of safetensors files first
|
|
|
|
files = list(path.glob("model-00001-of-*.safetensors"))
|
|
|
|
if not files:
|
|
|
|
# Try the PyTorch patterns too, with lower priority
|
2023-06-17 10:32:48 +00:00
|
|
|
globs = ["consolidated.00.pth", "pytorch_model-00001-of-*.bin", "*.pt", "pytorch_model.bin"]
|
2023-05-08 11:54:26 +00:00
|
|
|
files = [file for glob in globs for file in path.glob(glob)]
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
if not files:
|
|
|
|
raise Exception(f"Can't find model in directory {path}")
|
|
|
|
if len(files) > 1:
|
|
|
|
raise Exception(f"Found multiple models in {path}, not sure which to pick: {files}")
|
|
|
|
path = files[0]
|
|
|
|
|
|
|
|
paths = find_multifile_paths(path)
|
|
|
|
models_plus: List[ModelPlus] = []
|
|
|
|
for path in paths:
|
|
|
|
print(f"Loading model file {path}")
|
|
|
|
models_plus.append(lazy_load_file(path))
|
|
|
|
|
|
|
|
model_plus = merge_multifile_models(models_plus)
|
|
|
|
return model_plus
|
|
|
|
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
def load_vocab(path: Path, vocabtype: Optional[str]) -> Union[BpeVocab, SentencePieceVocab]:
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
# Be extra-friendly and accept either a file or a directory. Also, if it's
|
|
|
|
# a directory, it might be the model directory, and tokenizer.model might
|
|
|
|
# be in the parent of that.
|
|
|
|
if path.is_dir():
|
2023-07-25 13:22:09 +00:00
|
|
|
vocab_file = "tokenizer.model"
|
|
|
|
if vocabtype == 'bpe':
|
2023-08-21 20:07:43 +00:00
|
|
|
vocab_file = "vocab.json"
|
2023-07-25 13:22:09 +00:00
|
|
|
path2 = path / vocab_file
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
# Use `.parent` instead of /.. to handle the symlink case better.
|
2023-07-25 13:22:09 +00:00
|
|
|
path3 = path.parent / vocab_file
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
if path2.exists():
|
|
|
|
path = path2
|
|
|
|
elif path3.exists():
|
|
|
|
path = path3
|
|
|
|
else:
|
2023-06-17 10:32:48 +00:00
|
|
|
raise FileNotFoundError(
|
2023-08-22 18:58:16 +00:00
|
|
|
f"Could not find {vocab_file} in {path} or its parent; "
|
2023-06-17 10:32:48 +00:00
|
|
|
"if it's in another directory, pass the directory as --vocab-dir")
|
2023-08-21 20:07:43 +00:00
|
|
|
|
|
|
|
print(f"Loading vocab file '{path}', type '{vocabtype}'")
|
|
|
|
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
added_tokens_path = path.parent / "added_tokens.json"
|
2023-08-21 20:07:43 +00:00
|
|
|
if vocabtype == "bpe":
|
|
|
|
return BpeVocab(path, added_tokens_path if added_tokens_path.exists() else None)
|
|
|
|
elif vocabtype == "spm":
|
|
|
|
return SentencePieceVocab(path, added_tokens_path if added_tokens_path.exists() else None)
|
|
|
|
else:
|
|
|
|
raise ValueError(f"Unsupported vocabulary type {vocabtype}")
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
|
|
|
|
|
2023-06-22 12:20:47 +00:00
|
|
|
def default_outfile(model_paths: List[Path], file_type: GGMLFileType) -> Path:
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
namestr = {
|
2023-08-21 20:07:43 +00:00
|
|
|
GGMLFileType.AllF32: "f32",
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
GGMLFileType.MostlyF16: "f16",
|
2023-06-22 12:20:47 +00:00
|
|
|
}[file_type]
|
2023-08-21 20:07:43 +00:00
|
|
|
ret = model_paths[0].parent / f"ggml-model-{namestr}.gguf"
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
if ret in model_paths:
|
2023-06-17 10:32:48 +00:00
|
|
|
sys.stderr.write(
|
|
|
|
f"Error: Default output path ({ret}) would overwrite the input. "
|
|
|
|
"Please explicitly specify a path using --outfile.\n")
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
sys.exit(1)
|
|
|
|
return ret
|
|
|
|
|
|
|
|
|
|
|
|
def do_dump_model(model_plus: ModelPlus) -> None:
|
|
|
|
print(f"model_plus.paths = {model_plus.paths!r}")
|
|
|
|
print(f"model_plus.format = {model_plus.format!r}")
|
|
|
|
print(f"model_plus.vocab = {model_plus.vocab!r}")
|
|
|
|
for name, lazy_tensor in model_plus.model.items():
|
|
|
|
print(f"{name}: shape={lazy_tensor.shape} type={lazy_tensor.data_type}; {lazy_tensor.description}")
|
|
|
|
|
|
|
|
|
|
|
|
def main(args_in: Optional[List[str]] = None) -> None:
|
|
|
|
parser = argparse.ArgumentParser(description="Convert a LLaMa model to a GGML compatible file")
|
2023-08-21 20:07:43 +00:00
|
|
|
parser.add_argument("--dump", action="store_true", help="don't convert, just show what's in the model")
|
|
|
|
parser.add_argument("--dump-single", action="store_true", help="don't convert, just show what's in a single model file")
|
|
|
|
parser.add_argument("--vocab-only", action="store_true", help="extract only the vocab")
|
|
|
|
parser.add_argument("--outtype", choices=["f32", "f16"], help="output format (default: based on input)")
|
|
|
|
parser.add_argument("--vocab-dir", type=Path, help="directory containing tokenizer.model, if separate from model file")
|
|
|
|
parser.add_argument("--outfile", type=Path, help="path to write to; default: based on input")
|
|
|
|
parser.add_argument("model", type=Path, help="directory containing model file, or model file itself (*.pth, *.pt, *.bin)")
|
|
|
|
parser.add_argument("--vocabtype", choices=["spm", "bpe"], help="vocab format (default: spm)", default="spm")
|
|
|
|
parser.add_argument("--ctx", type=int, help="model training context (default: based on input)")
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
args = parser.parse_args(args_in)
|
|
|
|
|
|
|
|
if args.dump_single:
|
|
|
|
model_plus = lazy_load_file(args.model)
|
|
|
|
do_dump_model(model_plus)
|
2023-08-21 20:07:43 +00:00
|
|
|
|
|
|
|
model_plus = load_some_model(args.model)
|
|
|
|
|
|
|
|
params = Params.load(model_plus)
|
|
|
|
if params.n_ctx == -1:
|
|
|
|
if args.ctx is None:
|
|
|
|
raise Exception("The model doesn't have a context size, and you didn't specify one with --ctx\n"
|
|
|
|
"Please specify one with --ctx:\n"
|
|
|
|
" - LLaMA v1: --ctx 2048\n"
|
|
|
|
" - LLaMA v2: --ctx 4096\n")
|
|
|
|
params.n_ctx = args.ctx
|
|
|
|
|
2023-08-22 17:05:59 +00:00
|
|
|
if args.outtype:
|
|
|
|
params.ftype = {
|
|
|
|
"f32": GGMLFileType.AllF32,
|
|
|
|
"f16": GGMLFileType.MostlyF16,
|
|
|
|
}[args.outtype]
|
|
|
|
|
2023-08-21 20:07:43 +00:00
|
|
|
print(f"params = {params}")
|
|
|
|
|
|
|
|
vocab: Vocab
|
|
|
|
if args.vocab_only:
|
2023-07-25 13:22:09 +00:00
|
|
|
vocab = load_vocab(args.vocab_dir or args.model, args.vocabtype)
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
assert args.outfile, "need --outfile if using --vocab-only"
|
|
|
|
outfile = args.outfile
|
2023-08-21 20:07:43 +00:00
|
|
|
OutputFile.write_vocab_only(outfile, params, vocab)
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
print(f"Wrote {outfile}")
|
|
|
|
else:
|
|
|
|
if args.dump:
|
|
|
|
do_dump_model(model_plus)
|
|
|
|
return
|
2023-08-21 20:07:43 +00:00
|
|
|
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
if model_plus.vocab is not None and args.vocab_dir is None:
|
|
|
|
vocab = model_plus.vocab
|
|
|
|
else:
|
|
|
|
vocab_dir = args.vocab_dir if args.vocab_dir else model_plus.paths[0].parent
|
2023-07-25 13:22:09 +00:00
|
|
|
vocab = load_vocab(vocab_dir, args.vocabtype)
|
2023-08-21 20:07:43 +00:00
|
|
|
|
2023-08-22 17:05:59 +00:00
|
|
|
model = model_plus.model
|
|
|
|
model = convert_model_names(model, params)
|
|
|
|
ftype = pick_output_type(model, args.outtype)
|
|
|
|
model = convert_to_output_type(model, ftype)
|
|
|
|
outfile = args.outfile or default_outfile(model_plus.paths, ftype)
|
|
|
|
|
|
|
|
params.ftype = ftype
|
|
|
|
print(f"Writing {outfile}, format {ftype}")
|
2023-08-21 20:07:43 +00:00
|
|
|
|
|
|
|
OutputFile.write_all(outfile, params, model, vocab)
|
py : new conversion script (#545)
Current status: Working, except for the latest GPTQ-for-LLaMa format
that includes `g_idx`. This turns out to require changes to GGML, so
for now it only works if you use the `--outtype` option to dequantize it
back to f16 (which is pointless except for debugging).
I also included some cleanup for the C++ code.
This script is meant to replace all the existing conversion scripts
(including the ones that convert from older GGML formats), while also
adding support for some new formats. Specifically, I've tested with:
- [x] `LLaMA` (original)
- [x] `llama-65b-4bit`
- [x] `alpaca-native`
- [x] `alpaca-native-4bit`
- [x] LLaMA converted to 'transformers' format using
`convert_llama_weights_to_hf.py`
- [x] `alpaca-native` quantized with `--true-sequential --act-order
--groupsize 128` (dequantized only)
- [x] same as above plus `--save_safetensors`
- [x] GPT4All
- [x] stock unversioned ggml
- [x] ggmh
There's enough overlap in the logic needed to handle these different
cases that it seemed best to move to a single script.
I haven't tried this with Alpaca-LoRA because I don't know where to find
it.
Useful features:
- Uses multiple threads for a speedup in some cases (though the Python
GIL limits the gain, and sometimes it's disk-bound anyway).
- Combines split models into a single file (both the intra-tensor split
of the original and the inter-tensor split of 'transformers' format
files). Single files are more convenient to work with and more
friendly to future changes to use memory mapping on the C++ side. To
accomplish this without increasing memory requirements, it has some
custom loading code which avoids loading whole input files into memory
at once.
- Because of the custom loading code, it no longer depends in PyTorch,
which might make installing dependencies slightly easier or faster...
although it still depends on NumPy and sentencepiece, so I don't know
if there's any meaningful difference. In any case, I also added a
requirements.txt file to lock the dependency versions in case of any
future breaking changes.
- Type annotations checked with mypy.
- Some attempts to be extra user-friendly:
- The script tries to be forgiving with arguments, e.g. you can
specify either the model file itself or the directory containing
it.
- The script doesn't depend on config.json / params.json, just in
case the user downloaded files individually and doesn't have those
handy. But you still need tokenizer.model and, for Alpaca,
added_tokens.json.
- The script tries to give a helpful error message if
added_tokens.json is missing.
2023-04-14 07:03:03 +00:00
|
|
|
print(f"Wrote {outfile}")
|
|
|
|
|
|
|
|
|
|
|
|
if __name__ == '__main__':
|
|
|
|
main()
|