Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

add keyword extraction time elapsed to the little lamp. #3207

Merged
merged 1 commit into from
Nov 5, 2024
Merged

add keyword extraction time elapsed to the little lamp. #3207

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 6 additions & 1 deletion api/db/services/dialog_service.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -196,6 +196,8 @@ def chat(dialog, messages, stream=True, **kwargs):
questions = [full_question(dialog.tenant_id, dialog.llm_id, messages)]
else:
questions = questions[-1:]
refineQ_tm = timer()
keyword_tm = timer()

rerank_mdl = None
if dialog.rerank_id:
Expand All @@ -208,6 +210,7 @@ def chat(dialog, messages, stream=True, **kwargs):
else:
if prompt_config.get("keyword", False):
questions[-1] += keyword_extraction(chat_mdl, questions[-1])
keyword_tm = timer()

tenant_ids = list(set([kb.tenant_id for kb in kbs]))
kbinfos = retr.retrieval(" ".join(questions), embd_mdl, tenant_ids, dialog.kb_ids, 1, dialog.top_n,
Expand Down Expand Up @@ -267,7 +270,9 @@ def decorate_answer(answer):
if answer.lower().find("invalid key") >= 0 or answer.lower().find("invalid api") >= 0:
answer += " Please set LLM API-Key in 'User Setting -> Model Providers -> API-Key'"
done_tm = timer()
prompt += "\n\n### Elapsed\n - Retrieval: %.1f ms\n - LLM: %.1f ms"%((retrieval_tm-st)*1000, (done_tm-st)*1000)
prompt += "\n\n### Elapsed\n - Refine Question: %.1f ms\n - Keywords: %.1f ms\n - Retrieval: %.1f ms\n - LLM: %.1f ms" % (
(refineQ_tm - st) * 1000, (keyword_tm - refineQ_tm) * 1000, (retrieval_tm - keyword_tm) * 1000,
(done_tm - retrieval_tm) * 1000)
return {"answer": answer, "reference": refs, "prompt": prompt}

if stream:
Expand Down