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.

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

Chat: display failure text and restructure #1300

Merged
merged 5 commits into from
Jan 3, 2024

Conversation

rmackay9
Copy link
Contributor

@rmackay9 rmackay9 commented Jan 3, 2024

This includes a few unrelated changes:

  1. Chat window's status update displays a detailed "human readable" reason for Assistant failures. Eg. when usage limits are reached we receive this long message, "failed: Rate limit reached for gpt-4-1106-preview in organization org-fA312A0gDmmQRCAgojIFOi9G on tokens_usage_based per day: Limit 500000, Used 497355, Requested 4096. Please try again in 4m10.732s. Visit https://platform.openai.com/account/rate-limits to learn more." We could display a shorter "code" which would be just, "rate_limit_reached" but for now we display the full message
  2. Chat's function calls made consistent and redundant code removed by using getattr()
  3. Assistant Instructions improved for CONDITION_YAW to reduce mix-ups between relative and absolute angles
  4. Set-position-target-global-int simplifies function's "type_mask" field explanation simplified to improve chance that assistant picks the correct value

This has been tested in SITL.

@rmackay9 rmackay9 force-pushed the chat-display-failure-text branch from 3f53579 to 4527ce6 Compare January 3, 2024 03:00
@tridge tridge merged commit eacee97 into ArduPilot:master Jan 3, 2024
2 checks passed
@rmackay9 rmackay9 deleted the chat-display-failure-text branch January 3, 2024 06:59
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants