zhangyiss/mlx-examples
1
0
Fork 0
mirror of https://github.com/ml-explore/mlx-examples.git synced 2025-08-09 18:36:38 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Handle receiving 0 tokens gracefully (#231)

Browse Source 
* handle 0 tokens gracefully

* Formatting

* Move no token check to statistics section
This commit is contained in:
Christian Bieniak 2024-01-05 14:14:13 +11:00 committed by GitHub
parent 12c9bafbf5
commit 4fa659acbd
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 3 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
llms/hf_llm/generate.py
View File

@ -46,6 +46,9 @@ def generate(
print(tokenizer.decode(tokens)[skip:], flush=True) print(tokenizer.decode(tokens)[skip:], flush=True)
gen_time = time.time() - tic gen_time = time.time() - tic
print("=" * 10) print("=" * 10)
if len(tokens) == 0:
print("No tokens generated for this prompt")
return
prompt_tps = prompt.size / prompt_time prompt_tps = prompt.size / prompt_time
gen_tps = (len(tokens) - 1) / gen_time gen_tps = (len(tokens) - 1) / gen_time
print(f"Prompt: {prompt_tps:.3f} tokens-per-sec") print(f"Prompt: {prompt_tps:.3f} tokens-per-sec")