Mirrors/text-generation-webui
1
0
Fork 0
mirror of https://github.com/oobabooga/text-generation-webui.git synced 2024-09-20 18:45:09 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

Prevent *Is typing* from disappearing instantly while streaming

Browse source
This commit is contained in:
oobabooga 2023-03-15 12:51:13 -03:00
parent 4146ac4ff1
commit cf2da86352
1 changed files with 4 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
modules/text_generation.py
View file

@ -101,7 +101,8 @@ def generate_reply(question, max_new_tokens, do_sample, temperature, top_p, typi
reply = shared.model.generate(context=question, token_count=max_new_tokens, temperature=temperature, top_p=top_p, top_k=top_k) reply = shared.model.generate(context=question, token_count=max_new_tokens, temperature=temperature, top_p=top_p, top_k=top_k)
yield formatted_outputs(reply, shared.model_name) yield formatted_outputs(reply, shared.model_name)
else: else:
yield formatted_outputs(question, shared.model_name) if not (shared.args.chat or shared.args.cai_chat):
yield formatted_outputs(question, shared.model_name)
# 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_with_streaming(context=question, token_count=max_new_tokens, temperature=temperature, top_p=top_p, top_k=top_k): 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):
@ -197,7 +198,8 @@ def generate_reply(question, max_new_tokens, do_sample, temperature, top_p, typi
def generate_with_streaming(**kwargs): def generate_with_streaming(**kwargs):
return Iteratorize(generate_with_callback, kwargs, callback=None) return Iteratorize(generate_with_callback, kwargs, callback=None)
yield formatted_outputs(original_question, shared.model_name) if not (shared.args.chat or shared.args.cai_chat):
yield formatted_outputs(original_question, shared.model_name)
with generate_with_streaming(**generate_params) as generator: with generate_with_streaming(**generate_params) as generator:
for output in generator: for output in generator:
if shared.soft_prompt: if shared.soft_prompt: