Mirrors/text-generation-webui
1
0
Fork 0
mirror of https://github.com/oobabooga/text-generation-webui.git synced 2024-12-24 13:28:59 +01:00
Code Issues Packages Projects Releases Wiki Activity

Fix streaming

Browse Source
This commit is contained in:
oobabooga 2023-03-31 19:05:38 -03:00
parent 52065ae4cd
commit eeafd60713
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
modules/text_generation.py
View File

@ -130,7 +130,7 @@ def generate_reply(question, max_new_tokens, do_sample, temperature, top_p, typi
# RWKV has proper streaming, which is very nice. # RWKV has proper streaming, which is very nice.
# No need to generate 8 tokens at a time. # No need to generate 8 tokens at a time.
for reply in shared.model.generate(context=question, token_count=max_new_tokens, temperature=temperature, top_p=top_p, top_k=top_k, repetition_penalty=repetition_penalty): for reply in shared.model.generate_with_streaming(context=question, token_count=max_new_tokens, temperature=temperature, top_p=top_p, top_k=top_k, repetition_penalty=repetition_penalty):
output = original_question+reply output = original_question+reply
if not (shared.args.chat or shared.args.cai_chat): if not (shared.args.chat or shared.args.cai_chat):
reply = original_question + apply_extensions(reply, "output") reply = original_question + apply_extensions(reply, "output")