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

add --lowbit-path option for NPU llama example (#12020)

Browse source 
* add option" `--lowbit-path`

* add descriptions in `README.md` and formatting

* Update llama.py
This commit is contained in:
Ch1y0q 2024-09-05 15:31:01 +08:00 committed by GitHub
parent 8803242f5c
commit 820f8a4554
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 36 additions and 13 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

1
python/llm/example/NPU/HF-Transformers-AutoModels/LLM/README.md
View file

@ -118,6 +118,7 @@ python baichuan2.py
Arguments info:
- `--repo-id-or-model-path REPO_ID_OR_MODEL_PATH`: argument defining the huggingface repo id for the Llama2 model (i.e. `meta-llama/Llama-2-7b-chat-hf`) to be downloaded, or the path to the huggingface checkpoint folder. It is default to be `'meta-llama/Llama-2-7b-chat-hf'`.
- `--lowbit-path LOWBIT_MODEL_PATH`: argument defining the path to save/load lowbit version of the model. If it is an empty string, the original pretrained model specified by `REPO_ID_OR_MODEL_PATH` will be loaded. If it is an existing path, the lowbit model in `LOWBIT_MODEL_PATH` will be loaded. If it is a non-existing path, the original pretrained model specified by `REPO_ID_OR_MODEL_PATH` will be loaded, and the converted lowbit version will be saved into `LOWBIT_MODEL_PATH`. It is default to be `''`, i.e. an empty string.
- `--prompt PROMPT`: argument defining the prompt to be infered (with integrated prompt format for chat). It is default to be `What is AI?`.
- `--n-predict N_PREDICT`: argument defining the max number of tokens to predict. It is default to be `32`.
- `--max-output-len MAX_OUTPUT_LEN`: Defines the maximum sequence length for both input and output tokens. It is default to be `1024`.

48
python/llm/example/NPU/HF-Transformers-AutoModels/LLM/llama.py
View file

@ -50,6 +50,12 @@ if __name__ == "__main__":
help="The huggingface repo id for the Llama2 model to be downloaded"
", or the path to the huggingface checkpoint folder",
)
parser.add_argument("--lowbit-path", type=str,
default="",
help="The path to the lowbit model folder, leave blank if you do not want to save. \
If path not exists, lowbit model will be saved there. \
Else, lowbit model will be loaded.",
)
parser.add_argument('--prompt', type=str, default="What is AI?",
help='Prompt to infer')
parser.add_argument("--n-predict", type=int, default=32, help="Max tokens to predict")
@ -62,22 +68,38 @@ if __name__ == "__main__":
args = parser.parse_args()
model_path = args.repo_id_or_model_path
model = AutoModelForCausalLM.from_pretrained(
model_path,
torch_dtype=torch.float16,
trust_remote_code=True,
attn_implementation="eager",
load_in_low_bit="sym_int4",
optimize_model=True,
max_output_len=args.max_output_len,
max_prompt_len=args.max_prompt_len,
intra_pp=args.intra_pp,
inter_pp=args.inter_pp,
transpose_value_cache=not args.disable_transpose_value_cache,
)
if not args.lowbit_path or not os.path.exists(args.lowbit_path):
model = AutoModelForCausalLM.from_pretrained(
model_path,
torch_dtype=torch.float16,
trust_remote_code=True,
attn_implementation="eager",
load_in_low_bit="sym_int4",
optimize_model=True,
max_output_len=args.max_output_len,
max_prompt_len=args.max_prompt_len,
intra_pp=args.intra_pp,
inter_pp=args.inter_pp,
transpose_value_cache=not args.disable_transpose_value_cache,
)
else:
model = AutoModelForCausalLM.load_low_bit(
args.lowbit_path,
attn_implementation="eager",
torch_dtype=torch.float16,
optimize_model=True,
max_output_len=args.max_output_len,
max_prompt_len=args.max_prompt_len,
intra_pp=args.intra_pp,
inter_pp=args.inter_pp,
transpose_value_cache=not args.disable_transpose_value_cache,
)
tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
if args.lowbit_path and not os.path.exists(args.lowbit_path):
model.save_low_bit(args.lowbit_path)
DEFAULT_SYSTEM_PROMPT = """\
"""