fix: Excerpt from the thinking process #2181
Conversation
|
Adding the "do-not-merge/release-note-label-needed" label because no release-note block was detected, please follow our release note process to remove it. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here.
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
| if not self.reasoning_content_is_end: | ||
| self.reasoning_content_is_end = True | ||
| self.content += self.all_content | ||
| return {'content': self.all_content, 'reasoning_content': ''} | ||
| else: | ||
| if self.reasoning_content_is_start: | ||
| self.reasoning_content_chunk += chunk.content |
There was a problem hiding this comment.
There are no immediate issues with the provided code snippet. However, there is an unnecessary assignment within the if statement to modify self.reasoning_content_is_end.
I suggest optimizing it by removing that line since it unnecessarily increments the length of self.reasoning_content. Also, ensure that return {'content': '', 'reasoning_content': ''} is correctly placed within a conditional block where it makes sense to handle when both content parts have been fully collected.
Optimized version:
def get_reasoning_content(self, chunk):
if self.reasoning_content_is_start:
self.reasoning_content_chunk += chunk.content
elif not self.reasoning_content_is_complete:
# Check if we've reached the end tag
if 'end_of_reaso' in chunk.all_content.strip(): # Adjust condition based on actual tagging
self.reasoning_content_is_end = True
self.all_content += chunk.all_content
return {'content': "", 'reasoning_content': self.reasoning_content_chunk}
else: # Both chunks processed
return {'content': self.all_content, 'reasoning_content': self.reasoning_content_chunk}In this version, I assume that you meant to include additional logic to check for the end tag (end_of_reaso) and process the concatenated results appropriately. If this isn't the case, please adjust accordingly.
fix: Excerpt from the thinking process