98af5388ba
* bump flash attention 2.5.8 -> 2.6.1 * use triton implementation of cross entropy from flash attn * add smoke test for flash attn cross entropy patch * fix args to xentropy.apply * handle tuple from triton loss fn * ensure the patch tests run independently * use the wrapper already built into flash attn for cross entropy * mark pytest as forked for patches * use pytest xdist instead of forked, since cuda doesn't like forking * limit to 1 process and use dist loadfile for pytest * change up pytest for fixture to reload transformers w monkeypathc
46 lines
781 B
Plaintext
46 lines
781 B
Plaintext
--extra-index-url https://huggingface.github.io/autogptq-index/whl/cu118/
|
|
packaging==23.2
|
|
peft==0.11.1
|
|
transformers==4.42.3
|
|
tokenizers==0.19.1
|
|
bitsandbytes==0.43.1
|
|
accelerate==0.32.0
|
|
deepspeed @ git+https://github.com/microsoft/DeepSpeed.git@bc48371c5e1fb8fd70fc79285e66201dbb65679b
|
|
pydantic==2.6.3
|
|
addict
|
|
fire
|
|
PyYAML>=6.0
|
|
requests
|
|
datasets==2.19.1
|
|
flash-attn==2.6.1
|
|
sentencepiece
|
|
wandb
|
|
einops
|
|
xformers==0.0.27
|
|
optimum==1.16.2
|
|
hf_transfer
|
|
colorama
|
|
numba
|
|
numpy>=1.24.4
|
|
# qlora things
|
|
evaluate==0.4.1
|
|
scipy
|
|
scikit-learn==1.2.2
|
|
pynvml
|
|
art
|
|
fschat @ git+https://github.com/lm-sys/FastChat.git@27a05b04a35510afb1d767ae7e5990cbd278f8fe
|
|
gradio==3.50.2
|
|
tensorboard
|
|
python-dotenv==1.0.1
|
|
|
|
mamba-ssm==1.2.0.post1
|
|
|
|
# remote filesystems
|
|
s3fs
|
|
gcsfs
|
|
# adlfs
|
|
|
|
trl==0.9.6
|
|
zstandard==0.22.0
|
|
fastcore
|