Infer optimizer #7261
Open
Infer optimizer #7261
+118
−0
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
chenmoneygithub
requested changes
Jan 30, 2025
There was a problem hiding this comment.
@arnavsinghvi11 Interesting one! A few high-level comments:
- could you install ruff, and enable vscode/cursor auto formatting with ruff? There are many lint issues, but most of them can be automatically resolved.
- can we add some explanation on this optimizer with code examples? With a quick scan, I am wondering if this optimizer is only suitable for use cases where there are strong patterns.
| from dspy.evaluate.evaluate import Evaluate | ||
| import dspy | ||
|
|
||
| class BootstrapFewShotInfer(BootstrapFewShot): |
There was a problem hiding this comment.
Can we add docstring to explain what this optimizer do? like what's the key improvement from BTFS, and when this optimizer should be used.
| score, _ = evaluate(program, metric=self.metric) | ||
| return score | ||
|
|
||
| class RulesInductionProgram(dspy.Module): |
There was a problem hiding this comment.
maybe just call this RuleExtractor?
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
BootstrapFewShot defines the behavior behind the "how" in few-shot example selection by differentiating examples and selecting those that meet the metric criteria as "passing examples" in the optimized prompt.
BootstrapInfer takes a next step in exploring the "why" by inducing learned rules from the selected few-shot examples and adding supplementary guidance within the optimized prompt. The optimizer explores various sets of complementary rules + fewshot example combinations, and outputs a final optimal prompt with the original DSPy program instruction, selected few-shot examples and additional natural language rules to guide the task.