mmarch/text-generation-webui
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Change a comment

Browse source
This commit is contained in:
oobabooga 2023-02-18 22:58:40 -03:00 • committed by GitHub
parent 05e9da0c12
commit f79805f4a4
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
server.py
View file

@ -367,7 +367,7 @@ def generate_reply(question, tokens, do_sample, max_new_tokens, temperature, top
t1 = time.time() t1 = time.time()
print(f"Output generated in {(t1-t0):.2f} seconds ({(len(output)-len(input_ids[0]))/(t1-t0)/8:.2f} it/s, {len(output)-len(input_ids[0])} tokens)") print(f"Output generated in {(t1-t0):.2f} seconds ({(len(output)-len(input_ids[0]))/(t1-t0)/8:.2f} it/s, {len(output)-len(input_ids[0])} tokens)")
# Generate the reply 1 token at a time # Generate the reply 8 tokens at a time
else: else:
yield formatted_outputs(original_question, model_name) yield formatted_outputs(original_question, model_name)
for i in tqdm(range(tokens//8+1)): for i in tqdm(range(tokens//8+1)):