seqgra.learner.tensorflow.kerashelper module¶

MIT - CSAIL - Gifford Lab - seqgra

TensorFlow Keras learner helper class

@author: Konstantin Krismer

class KerasHelper[source]¶

Bases: object

MULTIPLE_REGRESSION_LOSSES: FrozenSet[str] = frozenset({'cosinesimilarity', 'huber', 'huberloss', 'mae', 'mape', 'meanabsoluteerror', 'meanabsolutepercentageerror', 'meansquarederror', 'meansquaredlogarithmicerror', 'mse', 'mslelogcosh', 'poisson'})¶

MULTIVARIATE_REGRESSION_LOSSES: FrozenSet[str] = frozenset({'cosinesimilarity', 'huber', 'huberloss', 'mae', 'mape', 'meanabsoluteerror', 'meanabsolutepercentageerror', 'meansquarederror', 'meansquaredlogarithmicerror', 'mse', 'mslelogcosh', 'poisson'})¶

MULTI_CLASS_CLASSIFICATION_LOSSES: FrozenSet[str] = frozenset({'categoricalcrossentropy', 'categoricalhinge', 'kld', 'kldivergence', 'kullbackleiblerdivergence', 'sparsecategoricalcrossentropy'})¶

MULTI_LABEL_CLASSIFICATION_LOSSES: FrozenSet[str] = frozenset({'binarycrossentropy', 'hinge', 'squaredhinge'})¶

static create_model(learner: seqgra.learner.learner.Learner) → None[source]¶

static evaluate_model(learner: seqgra.learner.learner.Learner, dataset)[source]¶

static get_keras_layer(operation)[source]¶

static get_loss(loss_hyperparameters)[source]¶

static get_num_params(learner: seqgra.learner.learner.Learner) → seqgra.schema.ModelSize [source]¶

static get_optimizer(optimizer_hyperparameters)[source]¶

static init_tf_memory_policy() → None[source]¶

static load_custom_weights(operation)[source]¶

static load_model(learner: seqgra.learner.learner.Learner, file_name: Optional[str] = None) → None[source]¶

static predict(learner: seqgra.learner.learner.Learner, dataset)[source]¶: This is the forward calculation from x to y :returns: Output tensor with the computed logits. :rtype: softmax_linear

static print_model_summary(learner: seqgra.learner.learner.Learner)[source]¶

static save_model(learner: seqgra.learner.learner.Learner, file_name: Optional[str] = None) → None[source]¶

static set_seed(learner: seqgra.learner.learner.Learner) → None[source]¶

static train_model(learner: seqgra.learner.learner.Learner, training_dataset, validation_dataset, silent: bool = False) → None[source]¶

static write_session_info(learner: seqgra.learner.learner.Learner) → None[source]¶