simpleml.models.classifiers.keras.seq2seq module¶

Seq2Seq Keras classifiers

class simpleml.models.classifiers.keras.seq2seq.KerasEncoderDecoderClassifier(save_method='disk_keras_hdf5', use_training_generator=False, training_generator_params={}, use_validation_generator=False, validation_generator_params={}, use_sequence_object=False, **kwargs)[source]¶

Bases: simpleml.models.classifiers.keras.seq2seq.KerasSeq2SeqClassifier

Specific subset of Seq2Seq models that contain encoder and decoder architectures

author¶

check_for_models(rebuild=False)[source]¶

created_timestamp¶

decode(X)[source]¶

encode(X)[source]¶

feature_metadata¶

filepaths¶

has_external_files¶

hash_¶

id¶

metadata_¶

modified_timestamp¶

name¶

params¶

pipeline¶

pipeline_id¶

project¶

registered_name¶

version¶

version_description¶

class simpleml.models.classifiers.keras.seq2seq.KerasEncoderDecoderStateClassifier(save_method='disk_keras_hdf5', use_training_generator=False, training_generator_params={}, use_validation_generator=False, validation_generator_params={}, use_sequence_object=False, **kwargs)[source]¶

Bases: simpleml.models.classifiers.keras.seq2seq.KerasEncoderDecoderClassifier

Specific subset of Seq2Seq models that contain encoder and decoder architectures with a state value to be propagated for each decoder timestep (eg LSTM/GRU decoder states)

author¶

created_timestamp¶

feature_metadata¶

filepaths¶

has_external_files¶

hash_¶

id¶

metadata_¶

modified_timestamp¶

name¶

params¶

pipeline¶

pipeline_id¶

project¶

registered_name¶

version¶

version_description¶

class simpleml.models.classifiers.keras.seq2seq.KerasEncoderDecoderStatelessClassifier(save_method='disk_keras_hdf5', use_training_generator=False, training_generator_params={}, use_validation_generator=False, validation_generator_params={}, use_sequence_object=False, **kwargs)[source]¶

Bases: simpleml.models.classifiers.keras.seq2seq.KerasEncoderDecoderStateClassifier

Specific subset of Seq2Seq models that contain encoder and decoder architectures withOUT a state value to be propagated for each decoder timestep. These architectures typically use repeat vectors to duplicate decoder inputs for later timesteps

author¶

created_timestamp¶

feature_metadata¶

filepaths¶

has_external_files¶

hash_¶

id¶

metadata_¶

modified_timestamp¶

name¶

params¶

pipeline¶

pipeline_id¶

project¶

registered_name¶

version¶

version_description¶

class simpleml.models.classifiers.keras.seq2seq.KerasSeq2SeqClassifier(save_method='disk_keras_hdf5', use_training_generator=False, training_generator_params={}, use_validation_generator=False, validation_generator_params={}, use_sequence_object=False, **kwargs)[source]¶

Bases: simpleml.models.classifiers.keras.model.KerasModelClassifier

Base class for sequence to sequence models. Differ from traditional models because training and inference use different architectures

author¶

build_inference_network(model)[source]¶

Inference network - Differs from training one so gets established dynamically at inference time

return: inference_model(s) rtype: self.external_model.__class__

check_for_models(rebuild=False)[source]¶

created_timestamp¶

feature_metadata¶

filepaths¶

has_external_files¶

hash_¶

id¶

metadata_¶

modified_timestamp¶

name¶

params¶

pipeline¶

pipeline_id¶

predict(X, **kwargs)[source]¶: Seq2Seq models have unpredictable results so overwrite batch process and return arrays instead of fixed size matrix (nXm vs nX1)

project¶

registered_name¶

version¶

version_description¶