2023-11-11 05:04:50 +00:00
|
|
|
#
|
|
|
|
# GGUF file reading/modification support. For API usage information,
|
|
|
|
# please see the files scripts/ for some fairly simple examples.
|
|
|
|
#
|
|
|
|
from __future__ import annotations
|
|
|
|
|
2024-05-03 19:36:41 +00:00
|
|
|
import logging
|
2023-11-11 05:04:50 +00:00
|
|
|
import os
|
|
|
|
from collections import OrderedDict
|
|
|
|
from typing import Any, Literal, NamedTuple, TypeVar, Union
|
|
|
|
|
|
|
|
import numpy as np
|
|
|
|
import numpy.typing as npt
|
|
|
|
|
2024-05-25 01:11:48 +00:00
|
|
|
from .quants import quant_shape_to_byte_shape
|
|
|
|
|
2023-11-11 05:04:50 +00:00
|
|
|
if __name__ == "__main__":
|
|
|
|
import sys
|
|
|
|
from pathlib import Path
|
|
|
|
|
|
|
|
# Allow running file in package as a script.
|
|
|
|
sys.path.insert(0, str(Path(__file__).parent.parent))
|
|
|
|
|
|
|
|
from gguf.constants import (
|
|
|
|
GGML_QUANT_SIZES,
|
|
|
|
GGUF_DEFAULT_ALIGNMENT,
|
|
|
|
GGUF_MAGIC,
|
|
|
|
GGUF_VERSION,
|
|
|
|
GGMLQuantizationType,
|
|
|
|
GGUFValueType,
|
|
|
|
)
|
|
|
|
|
2024-05-03 19:36:41 +00:00
|
|
|
logger = logging.getLogger(__name__)
|
2023-11-11 05:04:50 +00:00
|
|
|
|
|
|
|
READER_SUPPORTED_VERSIONS = [2, GGUF_VERSION]
|
|
|
|
|
|
|
|
|
|
|
|
class ReaderField(NamedTuple):
|
|
|
|
# Offset to start of this field.
|
|
|
|
offset: int
|
|
|
|
|
|
|
|
# Name of the field (not necessarily from file data).
|
|
|
|
name: str
|
|
|
|
|
|
|
|
# Data parts. Some types have multiple components, such as strings
|
|
|
|
# that consist of a length followed by the string data.
|
|
|
|
parts: list[npt.NDArray[Any]] = []
|
|
|
|
|
|
|
|
# Indexes into parts that we can call the actual data. For example
|
|
|
|
# an array of strings will be populated with indexes to the actual
|
|
|
|
# string data.
|
|
|
|
data: list[int] = [-1]
|
|
|
|
|
|
|
|
types: list[GGUFValueType] = []
|
|
|
|
|
|
|
|
|
|
|
|
class ReaderTensor(NamedTuple):
|
|
|
|
name: str
|
|
|
|
tensor_type: GGMLQuantizationType
|
|
|
|
shape: npt.NDArray[np.uint32]
|
|
|
|
n_elements: int
|
|
|
|
n_bytes: int
|
|
|
|
data_offset: int
|
|
|
|
data: npt.NDArray[Any]
|
|
|
|
field: ReaderField
|
|
|
|
|
|
|
|
|
|
|
|
class GGUFReader:
|
|
|
|
# I - same as host, S - swapped
|
2024-05-08 22:16:38 +00:00
|
|
|
byte_order: Literal['I'] | Literal['S'] = 'I'
|
2023-11-11 05:04:50 +00:00
|
|
|
alignment: int = GGUF_DEFAULT_ALIGNMENT
|
|
|
|
|
|
|
|
# Note: Internal helper, API may change.
|
|
|
|
gguf_scalar_to_np: dict[GGUFValueType, type[np.generic]] = {
|
|
|
|
GGUFValueType.UINT8: np.uint8,
|
|
|
|
GGUFValueType.INT8: np.int8,
|
|
|
|
GGUFValueType.UINT16: np.uint16,
|
|
|
|
GGUFValueType.INT16: np.int16,
|
|
|
|
GGUFValueType.UINT32: np.uint32,
|
|
|
|
GGUFValueType.INT32: np.int32,
|
|
|
|
GGUFValueType.FLOAT32: np.float32,
|
|
|
|
GGUFValueType.UINT64: np.uint64,
|
|
|
|
GGUFValueType.INT64: np.int64,
|
|
|
|
GGUFValueType.FLOAT64: np.float64,
|
|
|
|
GGUFValueType.BOOL: np.bool_,
|
|
|
|
}
|
|
|
|
|
2024-05-08 22:16:38 +00:00
|
|
|
def __init__(self, path: os.PathLike[str] | str, mode: Literal['r'] | Literal['r+'] | Literal['c'] = 'r'):
|
2023-11-11 05:04:50 +00:00
|
|
|
self.data = np.memmap(path, mode = mode)
|
|
|
|
offs = 0
|
|
|
|
if self._get(offs, np.uint32, override_order = '<')[0] != GGUF_MAGIC:
|
|
|
|
raise ValueError('GGUF magic invalid')
|
|
|
|
offs += 4
|
|
|
|
temp_version = self._get(offs, np.uint32)
|
|
|
|
if temp_version[0] & 65535 == 0:
|
|
|
|
# If we get 0 here that means it's (probably) a GGUF file created for
|
|
|
|
# the opposite byte order of the machine this script is running on.
|
|
|
|
self.byte_order = 'S'
|
|
|
|
temp_version = temp_version.newbyteorder(self.byte_order)
|
|
|
|
version = temp_version[0]
|
|
|
|
if version not in READER_SUPPORTED_VERSIONS:
|
|
|
|
raise ValueError(f'Sorry, file appears to be version {version} which we cannot handle')
|
|
|
|
self.fields: OrderedDict[str, ReaderField] = OrderedDict()
|
|
|
|
self.tensors: list[ReaderTensor] = []
|
|
|
|
offs += self._push_field(ReaderField(offs, 'GGUF.version', [temp_version], [0], [GGUFValueType.UINT32]))
|
|
|
|
temp_counts = self._get(offs, np.uint64, 2)
|
|
|
|
offs += self._push_field(ReaderField(offs, 'GGUF.tensor_count', [temp_counts[:1]], [0], [GGUFValueType.UINT64]))
|
|
|
|
offs += self._push_field(ReaderField(offs, 'GGUF.kv_count', [temp_counts[1:]], [0], [GGUFValueType.UINT64]))
|
|
|
|
tensor_count, kv_count = temp_counts
|
|
|
|
offs = self._build_fields(offs, kv_count)
|
|
|
|
offs, tensors_fields = self._build_tensors_fields(offs, tensor_count)
|
|
|
|
new_align = self.fields.get('general.alignment')
|
|
|
|
if new_align is not None:
|
2024-01-26 09:10:28 +00:00
|
|
|
if new_align.types != [GGUFValueType.UINT32]:
|
2023-11-11 05:04:50 +00:00
|
|
|
raise ValueError('Bad type for general.alignment field')
|
|
|
|
self.alignment = new_align.parts[-1][0]
|
|
|
|
padding = offs % self.alignment
|
|
|
|
if padding != 0:
|
|
|
|
offs += self.alignment - padding
|
|
|
|
self._build_tensors(offs, tensors_fields)
|
|
|
|
|
|
|
|
_DT = TypeVar('_DT', bound = npt.DTypeLike)
|
|
|
|
|
|
|
|
# Fetch a key/value metadata field by key.
|
|
|
|
def get_field(self, key: str) -> Union[ReaderField, None]:
|
|
|
|
return self.fields.get(key, None)
|
|
|
|
|
|
|
|
# Fetch a tensor from the list by index.
|
|
|
|
def get_tensor(self, idx: int) -> ReaderTensor:
|
|
|
|
return self.tensors[idx]
|
|
|
|
|
|
|
|
def _get(
|
2024-05-08 22:16:38 +00:00
|
|
|
self, offset: int, dtype: npt.DTypeLike, count: int = 1, override_order: None | Literal['I'] | Literal['S'] | Literal['<'] = None,
|
2023-11-11 05:04:50 +00:00
|
|
|
) -> npt.NDArray[Any]:
|
|
|
|
count = int(count)
|
|
|
|
itemsize = int(np.empty([], dtype = dtype).itemsize)
|
|
|
|
end_offs = offset + itemsize * count
|
|
|
|
return (
|
|
|
|
self.data[offset:end_offs]
|
|
|
|
.view(dtype = dtype)[:count]
|
|
|
|
.newbyteorder(override_order or self.byte_order)
|
|
|
|
)
|
|
|
|
|
|
|
|
def _push_field(self, field: ReaderField, skip_sum: bool = False) -> int:
|
|
|
|
if field.name in self.fields:
|
2024-04-29 13:58:41 +00:00
|
|
|
# TODO: add option to generate error on duplicate keys
|
|
|
|
# raise KeyError(f'Duplicate {field.name} already in list at offset {field.offset}')
|
|
|
|
|
2024-05-03 19:36:41 +00:00
|
|
|
logger.warning(f'Duplicate key {field.name} at offset {field.offset}')
|
2024-04-29 13:58:41 +00:00
|
|
|
self.fields[field.name + '_{}'.format(field.offset)] = field
|
|
|
|
else:
|
|
|
|
self.fields[field.name] = field
|
2023-11-11 05:04:50 +00:00
|
|
|
return 0 if skip_sum else sum(int(part.nbytes) for part in field.parts)
|
|
|
|
|
|
|
|
def _get_str(self, offset: int) -> tuple[npt.NDArray[np.uint64], npt.NDArray[np.uint8]]:
|
|
|
|
slen = self._get(offset, np.uint64)
|
|
|
|
return slen, self._get(offset + 8, np.uint8, slen[0])
|
|
|
|
|
|
|
|
def _get_field_parts(
|
|
|
|
self, orig_offs: int, raw_type: int,
|
|
|
|
) -> tuple[int, list[npt.NDArray[Any]], list[int], list[GGUFValueType]]:
|
|
|
|
offs = orig_offs
|
|
|
|
types: list[GGUFValueType] = []
|
|
|
|
gtype = GGUFValueType(raw_type)
|
|
|
|
types.append(gtype)
|
|
|
|
# Handle strings.
|
|
|
|
if gtype == GGUFValueType.STRING:
|
|
|
|
sparts: list[npt.NDArray[Any]] = list(self._get_str(offs))
|
|
|
|
size = sum(int(part.nbytes) for part in sparts)
|
|
|
|
return size, sparts, [1], types
|
|
|
|
# Check if it's a simple scalar type.
|
|
|
|
nptype = self.gguf_scalar_to_np.get(gtype)
|
|
|
|
if nptype is not None:
|
|
|
|
val = self._get(offs, nptype)
|
|
|
|
return int(val.nbytes), [val], [0], types
|
|
|
|
# Handle arrays.
|
|
|
|
if gtype == GGUFValueType.ARRAY:
|
|
|
|
raw_itype = self._get(offs, np.uint32)
|
|
|
|
offs += int(raw_itype.nbytes)
|
|
|
|
alen = self._get(offs, np.uint64)
|
|
|
|
offs += int(alen.nbytes)
|
|
|
|
aparts: list[npt.NDArray[Any]] = [raw_itype, alen]
|
|
|
|
data_idxs: list[int] = []
|
|
|
|
for idx in range(alen[0]):
|
|
|
|
curr_size, curr_parts, curr_idxs, curr_types = self._get_field_parts(offs, raw_itype[0])
|
|
|
|
if idx == 0:
|
|
|
|
types += curr_types
|
|
|
|
idxs_offs = len(aparts)
|
|
|
|
aparts += curr_parts
|
|
|
|
data_idxs += (idx + idxs_offs for idx in curr_idxs)
|
|
|
|
offs += curr_size
|
|
|
|
return offs - orig_offs, aparts, data_idxs, types
|
|
|
|
# We can't deal with this one.
|
|
|
|
raise ValueError('Unknown/unhandled field type {gtype}')
|
|
|
|
|
|
|
|
def _get_tensor(self, orig_offs: int) -> ReaderField:
|
|
|
|
offs = orig_offs
|
|
|
|
name_len, name_data = self._get_str(offs)
|
|
|
|
offs += int(name_len.nbytes + name_data.nbytes)
|
|
|
|
n_dims = self._get(offs, np.uint32)
|
|
|
|
offs += int(n_dims.nbytes)
|
|
|
|
dims = self._get(offs, np.uint64, n_dims[0])
|
|
|
|
offs += int(dims.nbytes)
|
|
|
|
raw_dtype = self._get(offs, np.uint32)
|
|
|
|
offs += int(raw_dtype.nbytes)
|
|
|
|
offset_tensor = self._get(offs, np.uint64)
|
|
|
|
offs += int(offset_tensor.nbytes)
|
|
|
|
return ReaderField(
|
|
|
|
orig_offs,
|
|
|
|
str(bytes(name_data), encoding = 'utf-8'),
|
|
|
|
[name_len, name_data, n_dims, dims, raw_dtype, offset_tensor],
|
|
|
|
[1, 3, 4, 5],
|
|
|
|
)
|
|
|
|
|
|
|
|
def _build_fields(self, offs: int, count: int) -> int:
|
|
|
|
for _ in range(count):
|
|
|
|
orig_offs = offs
|
|
|
|
kv_klen, kv_kdata = self._get_str(offs)
|
|
|
|
offs += int(kv_klen.nbytes + kv_kdata.nbytes)
|
|
|
|
raw_kv_type = self._get(offs, np.uint32)
|
|
|
|
offs += int(raw_kv_type.nbytes)
|
|
|
|
parts: list[npt.NDArray[Any]] = [kv_klen, kv_kdata, raw_kv_type]
|
|
|
|
idxs_offs = len(parts)
|
|
|
|
field_size, field_parts, field_idxs, field_types = self._get_field_parts(offs, raw_kv_type[0])
|
|
|
|
parts += field_parts
|
|
|
|
self._push_field(ReaderField(
|
|
|
|
orig_offs,
|
|
|
|
str(bytes(kv_kdata), encoding = 'utf-8'),
|
|
|
|
parts,
|
|
|
|
[idx + idxs_offs for idx in field_idxs],
|
|
|
|
field_types,
|
|
|
|
), skip_sum = True)
|
|
|
|
offs += field_size
|
|
|
|
return offs
|
|
|
|
|
|
|
|
def _build_tensors_fields(self, offs: int, count: int) -> tuple[int, list[ReaderField]]:
|
|
|
|
tensor_fields = []
|
|
|
|
for _ in range(count):
|
|
|
|
field = self._get_tensor(offs)
|
|
|
|
offs += sum(int(part.nbytes) for part in field.parts)
|
|
|
|
tensor_fields.append(field)
|
|
|
|
return offs, tensor_fields
|
|
|
|
|
|
|
|
def _build_tensors(self, start_offs: int, fields: list[ReaderField]) -> None:
|
|
|
|
tensors = []
|
2024-04-28 15:36:18 +00:00
|
|
|
tensor_names = set() # keep track of name to prevent duplicated tensors
|
2023-11-11 05:04:50 +00:00
|
|
|
for field in fields:
|
|
|
|
_name_len, name_data, _n_dims, dims, raw_dtype, offset_tensor = field.parts
|
2024-04-28 15:36:18 +00:00
|
|
|
# check if there's any tensor having same name already in the list
|
|
|
|
tensor_name = str(bytes(name_data), encoding = 'utf-8')
|
|
|
|
if tensor_name in tensor_names:
|
|
|
|
raise ValueError(f'Found duplicated tensor with name {tensor_name}')
|
|
|
|
tensor_names.add(tensor_name)
|
2023-11-11 05:04:50 +00:00
|
|
|
ggml_type = GGMLQuantizationType(raw_dtype[0])
|
2024-05-08 22:16:38 +00:00
|
|
|
n_elems = int(np.prod(dims))
|
2024-05-25 01:11:48 +00:00
|
|
|
np_dims = tuple(reversed(dims.tolist()))
|
2023-11-11 05:04:50 +00:00
|
|
|
block_size, type_size = GGML_QUANT_SIZES[ggml_type]
|
|
|
|
n_bytes = n_elems * type_size // block_size
|
|
|
|
data_offs = int(start_offs + offset_tensor[0])
|
|
|
|
item_type: npt.DTypeLike
|
gguf : add support for I64 and F64 arrays (#6062)
* gguf : add support for I64 and F64 arrays
GGML currently does not support I64 or F64 arrays and they are not often
used in machine learning, however if in the future the need arises, it
would be nice to add them now, so that the types are next to the other
types I8, I16, I32 in the enums, and it also reserves their type number.
Furthermore, with this addition the GGUF format becomes very usable for
most computational applications of NumPy (being compatible with the most
common NumPy dtypes: i8, i16, i32, i64, f32, f64), providing a faster,
and more versatile alternative to the `npz` format, and a simpler
alternative to the `hdf5` format.
The change in this PR seems small, not significantly increasing the
maintenance burden. I tested this from Python using GGUFWriter/Reader
and `gguf-dump`, as well as from C, everything seems to work.
* Fix compiler warnings
2024-03-15 08:46:51 +00:00
|
|
|
if ggml_type == GGMLQuantizationType.F16:
|
|
|
|
item_count = n_elems
|
|
|
|
item_type = np.float16
|
|
|
|
elif ggml_type == GGMLQuantizationType.F32:
|
2023-11-11 05:04:50 +00:00
|
|
|
item_count = n_elems
|
|
|
|
item_type = np.float32
|
gguf : add support for I64 and F64 arrays (#6062)
* gguf : add support for I64 and F64 arrays
GGML currently does not support I64 or F64 arrays and they are not often
used in machine learning, however if in the future the need arises, it
would be nice to add them now, so that the types are next to the other
types I8, I16, I32 in the enums, and it also reserves their type number.
Furthermore, with this addition the GGUF format becomes very usable for
most computational applications of NumPy (being compatible with the most
common NumPy dtypes: i8, i16, i32, i64, f32, f64), providing a faster,
and more versatile alternative to the `npz` format, and a simpler
alternative to the `hdf5` format.
The change in this PR seems small, not significantly increasing the
maintenance burden. I tested this from Python using GGUFWriter/Reader
and `gguf-dump`, as well as from C, everything seems to work.
* Fix compiler warnings
2024-03-15 08:46:51 +00:00
|
|
|
elif ggml_type == GGMLQuantizationType.F64:
|
2023-11-11 05:04:50 +00:00
|
|
|
item_count = n_elems
|
gguf : add support for I64 and F64 arrays (#6062)
* gguf : add support for I64 and F64 arrays
GGML currently does not support I64 or F64 arrays and they are not often
used in machine learning, however if in the future the need arises, it
would be nice to add them now, so that the types are next to the other
types I8, I16, I32 in the enums, and it also reserves their type number.
Furthermore, with this addition the GGUF format becomes very usable for
most computational applications of NumPy (being compatible with the most
common NumPy dtypes: i8, i16, i32, i64, f32, f64), providing a faster,
and more versatile alternative to the `npz` format, and a simpler
alternative to the `hdf5` format.
The change in this PR seems small, not significantly increasing the
maintenance burden. I tested this from Python using GGUFWriter/Reader
and `gguf-dump`, as well as from C, everything seems to work.
* Fix compiler warnings
2024-03-15 08:46:51 +00:00
|
|
|
item_type = np.float64
|
2024-03-14 10:40:14 +00:00
|
|
|
elif ggml_type == GGMLQuantizationType.I8:
|
|
|
|
item_count = n_elems
|
|
|
|
item_type = np.int8
|
|
|
|
elif ggml_type == GGMLQuantizationType.I16:
|
|
|
|
item_count = n_elems
|
|
|
|
item_type = np.int16
|
|
|
|
elif ggml_type == GGMLQuantizationType.I32:
|
|
|
|
item_count = n_elems
|
|
|
|
item_type = np.int32
|
gguf : add support for I64 and F64 arrays (#6062)
* gguf : add support for I64 and F64 arrays
GGML currently does not support I64 or F64 arrays and they are not often
used in machine learning, however if in the future the need arises, it
would be nice to add them now, so that the types are next to the other
types I8, I16, I32 in the enums, and it also reserves their type number.
Furthermore, with this addition the GGUF format becomes very usable for
most computational applications of NumPy (being compatible with the most
common NumPy dtypes: i8, i16, i32, i64, f32, f64), providing a faster,
and more versatile alternative to the `npz` format, and a simpler
alternative to the `hdf5` format.
The change in this PR seems small, not significantly increasing the
maintenance burden. I tested this from Python using GGUFWriter/Reader
and `gguf-dump`, as well as from C, everything seems to work.
* Fix compiler warnings
2024-03-15 08:46:51 +00:00
|
|
|
elif ggml_type == GGMLQuantizationType.I64:
|
|
|
|
item_count = n_elems
|
|
|
|
item_type = np.int64
|
2023-11-11 05:04:50 +00:00
|
|
|
else:
|
|
|
|
item_count = n_bytes
|
|
|
|
item_type = np.uint8
|
2024-05-25 01:11:48 +00:00
|
|
|
np_dims = quant_shape_to_byte_shape(np_dims, ggml_type)
|
2023-11-11 05:04:50 +00:00
|
|
|
tensors.append(ReaderTensor(
|
2024-04-28 15:36:18 +00:00
|
|
|
name = tensor_name,
|
2023-11-11 05:04:50 +00:00
|
|
|
tensor_type = ggml_type,
|
|
|
|
shape = dims,
|
|
|
|
n_elements = n_elems,
|
|
|
|
n_bytes = n_bytes,
|
|
|
|
data_offset = data_offs,
|
2024-05-25 01:11:48 +00:00
|
|
|
data = self._get(data_offs, item_type, item_count).reshape(np_dims),
|
2023-11-11 05:04:50 +00:00
|
|
|
field = field,
|
|
|
|
))
|
|
|
|
self.tensors = tensors
|