Skip to content

Correct handling of sample weights in FitGenetic #260

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

Open
gbrunin wants to merge 5 commits into
base: master
Choose a base branch
from
Open

Correct handling of sample weights in FitGenetic #260

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
1281977
Slightly relaxed tensorflow version.
gbrunin Oct 30, 2024
7c9c900
Fix pandas version.
gbrunin Mar 25, 2025
9e9186b
Merge branch 'master' of github.com:ppdebreuck/modnet
gbrunin Jul 21, 2025
5bf7bc3
Added sample_weight treatment in fit genetic.
gbrunin Aug 8, 2025
1dd1888
This fixes a bug that was triggered when early stopping was achieved,…
gbrunin Aug 8, 2025
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
12 changes: 12 additions & 0 deletions modnet/hyper_opt/fit_genetic.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -465,13 +465,21 @@ def function_fitness(
else:
n_splits = num_nested_folds
train_val_datas = []
sample_weights = []
for train, val in splits:
train_val_datas.append(self.train_data.split((train, val)))
if "sample_weight" in pop[0].fit_params:
sample_weights.append(pop[0].fit_params["sample_weight"][train])
else:
sample_weights.append(None)

tasks = []
for i, individual in enumerate(pop):
for j in range(n_splits):
train_data, val_data = train_val_datas[j]
sample_weight = sample_weights[j]
if sample_weight is not None:
individual.fit_params["sample_weight"] = sample_weight
tasks += [
{
"individual": individual,
Expand Down Expand Up @@ -653,6 +661,10 @@ def run(

self.best_model = EnsembleMODNetModel(models=ensemble)
"""
if "sample_weight" in fit_params:
self.best_individual.fit_params["sample_weight"] = fit_params[
"sample_weight"
]
self.best_model = self.best_individual.refit_model(
self.data, n_models=refit, n_jobs=n_jobs or 1, fast=fast
)
Expand Down