nirs4all.operators.models.tensorflow.nicon module

nirs4all.operators.models.tensorflow.nicon.customizable_decon(input_shape, params=None)

Builds a model using depthwise separable convolutions and layer normalization.

Parameters:

• input_shape (tuple) – Shape of the input data.

• params (dict) – Dictionary of parameters for model configuration.

Returns:

Compiled deconvolutional model.

Return type:

keras.Model

nirs4all.operators.models.tensorflow.nicon.customizable_decon_classification(input_shape, num_classes=2, params=None)

Builds a customizable model using depthwise separable convolutions for classification.

Parameters:

• input_shape (tuple) – Shape of the input data.

• num_classes (int) – Number of classes for classification.

• params (dict) – Dictionary of parameters for model configuration.

Returns:

Compiled classification model.

Return type:

keras.Model

nirs4all.operators.models.tensorflow.nicon.customizable_nicon(input_shape, params=None)

Builds a custom CNN model with depthwise convolutions.

Parameters:

• input_shape (tuple) – Shape of the input data.

• params (dict) – Dictionary of parameters for model configuration.

Returns:

Compiled model.

Return type:

keras.Sequential

nirs4all.operators.models.tensorflow.nicon.customizable_nicon_classification(input_shape, num_classes=2, params=None)

Builds a custom CNN model with depthwise convolutions for classification.

Parameters:

• input_shape (tuple) – Shape of the input data.

• num_classes (int) – Number of classes for classification.

• params (dict) – Dictionary of parameters for model configuration.

Returns:

Compiled classification model.

Return type:

keras.Sequential

nirs4all.operators.models.tensorflow.nicon.decon(input_shape, params=None)

Builds a CNN model with depthwise and separable convolutions.

Parameters:

• input_shape (tuple) – Shape of the input data.

• params (dict) – Dictionary of parameters for model configuration.

Returns:

Compiled model.

Return type:

keras.Sequential

nirs4all.operators.models.tensorflow.nicon.decon_Sep(input_shape, params=None)

Builds a CNN model with separable convolutions.

Parameters:

• input_shape (tuple) – Shape of the input data.

• params (dict) – Dictionary of parameters for model configuration.

Returns:

Compiled model.

Return type:

keras.Sequential

nirs4all.operators.models.tensorflow.nicon.decon_Sep_classification(input_shape, num_classes=2, params=None)

Builds a CNN model with separable convolutions for classification.

Parameters:

• input_shape (tuple) – Shape of the input data.

• num_classes (int) – Number of classes for classification.

• params (dict) – Dictionary of parameters for model configuration.

Returns:

Compiled classification model.

Return type:

keras.Sequential

nirs4all.operators.models.tensorflow.nicon.decon_classification(input_shape, num_classes=2, params=None)

Builds a CNN model with depthwise and separable convolutions for classification.

Parameters:

• input_shape (tuple) – Shape of the input data.

• num_classes (int) – Number of classes for classification.

• params (dict) – Dictionary of parameters for model configuration.

Returns:

Compiled classification model.

Return type:

keras.Sequential

nirs4all.operators.models.tensorflow.nicon.nicon(input_shape, params=None)

Builds a custom CNN model with depthwise convolutions.

Parameters:

• input_shape (tuple) – Shape of the input data.

• params (dict) – Dictionary of parameters for model configuration.

Returns:

Compiled model.

Return type:

keras.Sequential

nirs4all.operators.models.tensorflow.nicon.nicon_VG(input_shape, params=None)

Builds a custom CNN model.

Parameters:

• input_shape (tuple) – Shape of the input data.

• params (dict) – Dictionary of parameters for model configuration.

Returns:

Compiled CNN model.

Return type:

keras.Sequential

nirs4all.operators.models.tensorflow.nicon.nicon_VG_classification(input_shape, num_classes=2, params=None)

Builds a custom CNN model for classification.

Parameters:

• input_shape (tuple) – Shape of the input data.

• num_classes (int) – Number of classes for classification.

• params (dict) – Dictionary of parameters for model configuration.

Returns:

Compiled classification model.

Return type:

keras.Sequential

nirs4all.operators.models.tensorflow.nicon.nicon_classification(input_shape, num_classes=2, params=None)

Builds a custom CNN model with depthwise convolutions for classification.

Parameters:

• input_shape (tuple) – Shape of the input data.

• num_classes (int) – Number of classes for classification.

• params (dict) – Dictionary of parameters for model configuration.

Returns:

Compiled classification model.

Return type:

keras.Sequential

nirs4all.operators.models.tensorflow.nicon.thin_nicon(input_shape, params=None)

Builds a custom CNN model with depthwise convolutions.

Parameters:

• input_shape (tuple) – Shape of the input data.

• params (dict) – Dictionary of parameters for model configuration.

Returns:

Compiled model.

Return type:

keras.Sequential

nirs4all.operators.models.tensorflow.nicon.transformer(input_shape, params=None)

nirs4all.operators.models.tensorflow.nicon.transformer_VG(input_shape, params=None)

nirs4all.operators.models.tensorflow.nicon.transformer_VG_classification(input_shape, num_classes=2, params=None)

nirs4all.operators.models.tensorflow.nicon.transformer_classification(input_shape, num_classes=2, params=None)

nirs4all.operators.models.tensorflow.nicon.transformer_model(input_shape, params=None)

Builds a transformer model for 1D data.

Parameters:

• input_shape (tuple) – Shape of the input data.

• params (dict) – Dictionary of parameters for model configuration.

Returns:

Compiled transformer model.

Return type:

keras.Model

nirs4all.operators.models.tensorflow.nicon.transformer_model_classification(input_shape, num_classes=2, params=None)

Builds a transformer model for 1D data classification.

Parameters:

• input_shape (tuple) – Shape of the input data.

• num_classes (int) – Number of classes for classification.

• params (dict) – Dictionary of parameters for model configuration.

Returns:

Compiled transformer classification model.

Return type:

keras.Model