Vanilla-pad baseline does not run

Closed Issue created 3 years ago

Following the documentation, the vanilla-pad baseline can be invoked with:

bob pad vanilla-pad replay-attack svm-frames -o results -vv

This leads to the following exception:

Traceback (most recent call last):
  File "/remote/idiap.svm/temp.biometric03/cecabert/bob_beta/src/bob.pipelines/bob/pipelines/wrappers.py", line 808, in _fit
    self.estimator = self.estimator.fit(X, y, **fit_params)
  File "/remote/idiap.svm/temp.biometric03/cecabert/bob_beta/src/bob.pipelines/bob/pipelines/wrappers.py", line 337, in fit
    self.estimator = self.estimator.fit(X, **kwargs)
  File "/remote/idiap.svm/temp.biometric03/cecabert/mambaforge/envs/bob_deps/lib/python3.10/site-packages/sklearn/svm/_base.py", line 190, in fit
    X, y = self._validate_data(
  File "/remote/idiap.svm/temp.biometric03/cecabert/mambaforge/envs/bob_deps/lib/python3.10/site-packages/sklearn/base.py", line 581, in _validate_data
    X, y = check_X_y(X, y, **check_params)
  File "/remote/idiap.svm/temp.biometric03/cecabert/mambaforge/envs/bob_deps/lib/python3.10/site-packages/sklearn/utils/validation.py", line 964, in check_X_y
    X = check_array(
  File "/remote/idiap.svm/temp.biometric03/cecabert/mambaforge/envs/bob_deps/lib/python3.10/site-packages/sklearn/utils/validation.py", line 794, in check_array
    raise ValueError(
ValueError: Found array with dim 4. Estimator expected <= 2.

The fit method of the SVM classifier is expecting its input to be (n_samples, n_features) which is not what is provided by the VideoToFrames transformer. One possible solution would be to add an intermediate step between the two operations to flatten the data.

Vanilla-pad baseline does not run

Linked items ... 0

Activity