ayo/ipex-llm
1
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

fix fschat DEP version error (#10325)

Browse source
This commit is contained in:
ZehuaCao 2024-03-06 16:15:27 +08:00 committed by GitHub
parent 9ea499ca68
commit 267de7abc3
2 changed files with 6 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
docker/llm/serving/cpu/docker/model_adapter.py.patch
View file

@ -1,11 +1,11 @@
--- model_adapter.py.old 2024-01-24 01:56:23.903144335 +0000 --- model_adapter.py.old 2024-03-05 15:08:47.169275336 +0800
+++ model_adapter.py 2024-01-24 01:59:22.605062765 +0000 +++ model_adapter.py 2024-03-05 15:10:13.434703674 +0800
@@ -1346,15 +1346,17 @@ @@ -1690,15 +1690,17 @@
) )
# NOTE: if you use the old version of model file, please remove the comments below # NOTE: if you use the old version of model file, please remove the comments below
# config.use_flash_attn = False # config.use_flash_attn = False
- config.fp16 = True - self.float_set(config, "fp16")
+ # config.fp16 = True + # self.float_set(config, "fp16")
generation_config = GenerationConfig.from_pretrained( generation_config = GenerationConfig.from_pretrained(
model_path, trust_remote_code=True model_path, trust_remote_code=True
) )

2
python/llm/setup.py
View file

@ -53,7 +53,7 @@ CONVERT_DEP = ['numpy >= 1.22', 'torch',
'transformers == 4.31.0', 'sentencepiece', 'tokenizers == 0.13.3', 'transformers == 4.31.0', 'sentencepiece', 'tokenizers == 0.13.3',
# TODO: Support accelerate 0.22.0 # TODO: Support accelerate 0.22.0
'accelerate == 0.21.0', 'tabulate'] 'accelerate == 0.21.0', 'tabulate']
SERVING_DEP = ['fschat[model_worker, webui] == 0.2.28', 'protobuf'] SERVING_DEP = ['fschat[model_worker, webui] == 0.2.36', 'protobuf']
windows_binarys = [ windows_binarys = [
"llama.dll", "llama.dll",
"gptneox.dll", "gptneox.dll",