:param y_true: (batch, lenght, input_size, 2) :param y_pred: (batch, lenght, input_size, 2) :return:

:param tensor: :return:

y_pred has np nan where we shouldn't compute the loss :param y_true: (batch, nb_steps=1, step_size, input_size, channels=1)

:param y_true: (batch, nb_steps, step_length, input_size, channels) :param y_pred: (batch, nb_steps, step_length, input_size, channe

:param l: :return:

:param y_true: (batch, nb_instruments, nb_steps, step_size, input_size, channels) :param y_pred: (batch, nb_instruments, nb_steps, s

:param y_true: (batch, nb_steps=1, step_size, input_size, channels=1) :param y_pred: (batch, nb_steps=1, step_size, input_size, chann

:param y_true: :param y_pred: :return:

:param y_true: (batch, nb_steps=1, step_size, input_size, channels=1) :param y_pred: (batch, nb_steps=1, step_size, input_size, chann

:param y_true: (batch, nb_steps=1, step_size, input_size, channels=1) :param y_pred: (batch, nb_steps=1, step_size, input_size, chann

Used to add the weights of a sublayers in a model Since Tensorflow 2.0 it works without this function :param args:

:param path: the root path :param small_data: if we want to keep only a small amount of data :return: An array of all the path of all th

:param layers: :return:

:param layers: :return:

:param layer: :return:

Print the Git hash of this version of `diagnose_tensorboard.py`. Given this hash, use `git cat-file blob HASH` to recover the relevant versio

Preprocess the args for the file bayesian-opt.py :param args: :return:

:param input_shape: [(None, (None?), in_features), adj, labels]

:param input_shape: :return:

:return:

:param input_shape: List(2)[(batch, nb_instruments, nb_steps, size)] + List(1)[(batch, nb_instrument, nb_steps)] :return:

:param input_shape: (?, input_size) :return:

:param input_shape: (?, previous_size) :return:

:param input_shape: (?, previous_size) :return:

:param input_shape: (batch, step_length, size, channels=1) :return:

:param input_shape: (batch, nb_steps, input_size) :return:

:param inputs_shape: if not bidirectional: [(batch, nb_units)] or (batch, nb_units)

:param inputs_shape: (batch, size) :return:

:param inputs: [(None, (None?), in_features), adj, labels] :return: (None, (None?), out_features)

:param inputs: :return:

:param inputs: List(2)[(batch, modalities, size)] ([means, std]) :return:

:param inputs: List(2)[(batch, modalities, nb_steps?, size)] + List(1)[(batch, modalities, nb_steps?)] ([means, std]) :r

:param inputs: List(2)[(batch, modalities, nb_steps, size)] + List(1)[(batch, modalities, nb_steps)] ([means, std]) :ret

:param x: List(2)[(batch, nb_instruments, nb_steps, size)] + List(1)[(batch, nb_instrument, nb_steps)] :return:

:param inputs: List[tensor] :return: