Use structured prompt for all model call
What does this merge request do and why?
This MR uses a structured prompt instead of just a simple string during LLM calls. This change is necessary to fully utilize Anthropic's message API (also OpenAI's API)
- The components of the structured prompt is read from different templates
- Move prompt template input tuple out of prompt-engine sub module. This is much more readable.
- Apply the components of the structured prompt in the correct place for anthropic and openai models.
Ref: #180 #190
How to set up and validate locally
Numbered steps to set up and validate the change are strongly suggested.
Merge request checklist
-
I've ran the affected pipeline(s) to validate that nothing is broken. -
Tests added for new functionality. If not, please raise an issue to follow up. -
Documentation added/updated, if needed.
Edited by Hongtao Yang