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.

Sign up for GitHub

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

fix(sweeps): actually use minimum_improvement value #114

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

fix(sweeps): actually use minimum_improvement value #114

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions src/sweeps/bayes_search.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -200,7 +200,7 @@ def next_sample(
maximum samples to consider (since algo is O(n^3)) for performance,
but also adds some randomness. this number of samples will be chosen
randomly from the sample_X and used to train the GP.
improvement: floating, optional, default 0.1
improvement: floating, optional, default 0.01
amount of improvement to optimize for -- higher means take more exploratory risks
num_points_to_try: integer, optional, default 1000
number of X values to try when looking for value with highest expected probability
Expand Down Expand Up @@ -288,7 +288,7 @@ def next_sample(
min_norm_y = (min_unnorm_y - y_mean) / y_stddev - improvement
else:
"""
min_norm_y = (min_unnorm_y - y_mean) / y_stddev
min_norm_y = (min_unnorm_y - y_mean) / y_stddev - improvement

Z = -(y_pred - min_norm_y) / (y_pred_std + STD_NUMERICAL_STABILITY_EPSILON)
prob_of_improve: np.ndarray = scipy_stats.norm.cdf(Z)
Expand Down