github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-06-25 16:13:25 +00:00
Code Issues Packages Projects Releases Wiki Activity

DOCS: Fix typo/line break in python code (#13708)

Browse Source
This commit is contained in:
h3l 2023-11-23 01:10:07 +08:00 committed by GitHub
parent 5b90fe5b1c
commit 14d4fb98fc
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 9 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
templates/chat-bot-feedback/README.md
View File

@ -26,8 +26,7 @@ As shown, the evaluator sees that the user is increasingly frustrated, indicatin
The user feedback is inferred by custom `RunEvaluator`. This evaluator is called using the `EvaluatorCallbackHandler`, which run it in a separate thread to avoid interfering with the chat bot's runtime. You can use this custom evaluator on any compatible chat bot by calling the following function on your LangChain object: The user feedback is inferred by custom `RunEvaluator`. This evaluator is called using the `EvaluatorCallbackHandler`, which run it in a separate thread to avoid interfering with the chat bot's runtime. You can use this custom evaluator on any compatible chat bot by calling the following function on your LangChain object:
```python ```python
my_chain my_chain.with_config(
.with_config(
callbacks=[ callbacks=[
EvaluatorCallbackHandler( EvaluatorCallbackHandler(
evaluators=[ evaluators=[
@ -35,7 +34,7 @@ my_chain
] ]
) )
], ],
) )
``` ```
The evaluator instructs an LLM, specifically `gpt-3.5-turbo`, to evaluate the AI's most recent chat message based on the user's followup response. It generates a score and accompanying reasoning that is converted to feedback in LangSmith, applied to the value provided as the `last_run_id`. The evaluator instructs an LLM, specifically `gpt-3.5-turbo`, to evaluate the AI's most recent chat message based on the user's followup response. It generates a score and accompanying reasoning that is converted to feedback in LangSmith, applied to the value provided as the `last_run_id`.