Resnet classifier

polars_ts.classification.resnet_classifier

ResNet classifier for time series.

Implements a 1D ResNet adapted for time series classification with residual blocks and global average pooling.

References

Wang et al. (2017). Time Series Classification from Scratch with Deep Neural Networks: A Strong Baseline. IJCNN.

_ResBlock

Bases: Module

1D residual block.

_ResNet1D

Bases: Module

1D ResNet for time series classification.

ResNetClassifier

1D ResNet time series classifier.

Parameters

n_filters Base number of convolutional filters. max_epochs Maximum training epochs. lr Learning rate. batch_size Training batch size. id_col, target_col, time_col Column names.

fit(df, *, label_col='label')

Train on labeled time series data.

predict(df)

Predict class labels for test series.