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

fix chatglm in run.pu (#8919)

Browse source
This commit is contained in:
Xin Qiu 2023-09-07 16:44:10 +08:00 committed by GitHub
parent e9de9d9950
commit d8a01d7c4f
1 changed files with 2 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
python/llm/dev/benchmark/all-in-one/run.py
View file

@ -127,6 +127,8 @@ def run_transformer_int4(repo_id,
# slice the input_ids to ensure the prompt length is required length. # slice the input_ids to ensure the prompt length is required length.
input_ids = tokenizer.encode(input_str, return_tensors="pt") input_ids = tokenizer.encode(input_str, return_tensors="pt")
input_ids = input_ids[:, :in_len] input_ids = input_ids[:, :in_len]
true_str = tokenizer.batch_decode(input_ids)[0]
input_ids = tokenizer.encode(true_str, return_tensors="pt")
result[in_out] = [] result[in_out] = []
for i in range(num_trials + warm_up): for i in range(num_trials + warm_up):
st = time.perf_counter() st = time.perf_counter()