Skip to content

armgit / dspy

Public Imported from GitHub
0

DSPy: The framework for programming—not prompting—language models

Python

Support for max_errors in randomsearch

Thomas D Ahle3069b4ec5da3
diff --git a/dspy/teleprompt/random_search.py b/dspy/teleprompt/random_search.py
index 8605cd427..43d2c5d4c 100644
--- a/dspy/teleprompt/random_search.py
+++ b/dspy/teleprompt/random_search.py
@@ -28,7 +28,7 @@ from dspy.evaluate.evaluate import Evaluate
class BootstrapFewShotWithRandomSearch(Teleprompter):
- def __init__(self, metric, teacher_settings={}, max_bootstrapped_demos=4, max_labeled_demos=16, max_rounds=1, num_candidate_programs=16, num_threads=6, stop_at_score=None):
+ def __init__(self, metric, teacher_settings={}, max_bootstrapped_demos=4, max_labeled_demos=16, max_rounds=1, num_candidate_programs=16, num_threads=6, max_errors=10, stop_at_score=None):
self.metric = metric
self.teacher_settings = teacher_settings
self.max_rounds = max_rounds
@@ -37,6 +37,7 @@ class BootstrapFewShotWithRandomSearch(Teleprompter):
self.stop_at_score = stop_at_score
self.min_num_samples = 1
self.max_num_samples = max_bootstrapped_demos
+ self.max_erros = max_errors
self.num_candidate_sets = num_candidate_programs
# self.max_num_traces = 1 + int(max_bootstrapped_demos / 2.0 * self.num_candidate_sets)
@@ -93,7 +94,7 @@ class BootstrapFewShotWithRandomSearch(Teleprompter):
program2 = teleprompter.compile(student, teacher=teacher, trainset=trainset2)
evaluate = Evaluate(devset=self.valset, metric=self.metric, num_threads=self.num_threads,
- display_table=False, display_progress=True)
+ max_errors=self.max_errors, display_table=False, display_progress=True)
score, subscores = evaluate(program2, return_all_scores=True)