root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-26 03:14:35 +00:00
Code Issues Actions 7 Packages Projects Releases Wiki Activity
81bc9214a3
llama.cpp/examples/train-text-from-scratch
History
Hongyu Ouyang 81bc9214a3
train : fix #4227 (double free in examples/train-text-from-scratch/train-text-from-scratch.cpp) (#4351) 
On commit b1108 (44c117f4) xaedes added

    ggml_allocr * alloc = NULL;

    ... (many lines in between)

    if (alloc) {
        ggml_allocr_free(alloc);
    }

Which is correct, but it's easy to lose context after many lines in between.

On commit b1287 (0e76a899) xaedes made a big change. From here on, alloc is freed eagerly.

    alloc = ggml_allocr_new(...)
    ... (short lines of code)
    ggml_allocr_free(alloc)

This happens a few times, but alloc is never set to NULL, and many lines below,
we still have

    if (alloc) {
        ggml_allocr_free(alloc);
    }

which causes a double-free.
2023-12-07 12:25:22 +02:00
..
CMakeLists.txt cmake : install targets (#2256) 2023-07-19 10:01:11 +03:00
convert-train-checkpoint-to-gguf.py gguf-py: Refactor and allow reading/modifying existing GGUF files (#3981) 2023-11-11 08:04:50 +03:00
README.md train : finetune LORA (#2632) 2023-09-28 21:40:11 +03:00
train-text-from-scratch.cpp train : fix #4227 (double free in examples/train-text-from-scratch/train-text-from-scratch.cpp) (#4351) 2023-12-07 12:25:22 +02:00

README.md

train-text-from-scratch

Basic usage instructions:

# get training data
wget https://raw.githubusercontent.com/brunoklein99/deep-learning-notes/master/shakespeare.txt

# train
./bin/train-text-from-scratch \
        --vocab-model ../models/ggml-vocab-llama.gguf \
        --ctx 64 --embd 256 --head 8 --layer 16 \
        --checkpoint-in  chk-shakespeare-256x16-LATEST.gguf \
        --checkpoint-out chk-shakespeare-256x16-ITERATION.gguf \
        --model-out ggml-shakespeare-256x16-f32-ITERATION.gguf \
        --train-data "shakespeare.txt" \
        -t 6 -b 16 --seed 1 --adam-iter 256 \
        --no-checkpointing

# predict
./bin/main -m ggml-shakespeare-256x16-f32.gguf

Output files will be saved every N iterations (config with --save-every N). The pattern "ITERATION" in the output filenames will be replaced with the iteration number and "LATEST" for the latest output.

To train GGUF models just pass them to --checkpoint-in FN.