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鈥檒l occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Optimize find_best_temp_scaler for performance #1075

Open
wants to merge 3 commits into
base: master
Choose a base branch
from
Open

Optimize find_best_temp_scaler for performance #1075

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
e855b27
Perf: compute_soft_cross_entropy
gogetron Mar 30, 2024
80a5eb8
Fix: remove unused import
gogetron Mar 30, 2024
44c4329
Merge branch 'master' into perf-find_best_temp_scaler
gogetron May 26, 2024
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
17 changes: 10 additions & 7 deletions cleanlab/internal/multiannotator_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,7 +25,7 @@
import pandas as pd

from cleanlab.internal.numerics import softmax
from cleanlab.internal.util import get_num_classes, value_counts
from cleanlab.internal.util import get_num_classes
from cleanlab.internal.validation import assert_valid_class_labels
from cleanlab.typing import LabelLike

Expand Down Expand Up @@ -275,12 +275,15 @@ def compute_soft_cross_entropy(
"""Compute soft cross entropy between the annotators' empirical label distribution and model pred_probs"""
num_classes = get_num_classes(pred_probs=pred_probs)

empirical_label_distribution = np.full((len(labels_multiannotator), num_classes), np.NaN)
for i, labels in enumerate(labels_multiannotator):
labels_subset = labels[~np.isnan(labels)]
empirical_label_distribution[i, :] = value_counts(
labels_subset, num_classes=num_classes
) / len(labels_subset)
empirical_label_distribution = np.zeros((len(labels_multiannotator), num_classes), dtype=float)
length = np.zeros(len(labels_multiannotator), dtype=float)
for i in range(labels_multiannotator.shape[1]):
mask = ~np.isnan(labels_multiannotator[:, i])
empirical_label_distribution[mask, labels_multiannotator[mask, i].astype(int)] += 1
length += mask

for k in range(num_classes):
empirical_label_distribution[:, k] /= length

clipped_pred_probs = np.clip(pred_probs, a_min=SMALL_CONST, a_max=None)
soft_cross_entropy = -np.sum(
Expand Down
Loading