Setup the optimizer. We provide a reasonable default that works well. If you want to use something else, you can pass a tuple in the

r""" Args: labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): Labels for computing

r""" labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for computing the sequence classification/regression

r""" labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for computing the sequence classification/regression

r""" start_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for position (index) of the start of the la

Forward function.

Compute scaled dot product attention. Args: query (torch.Tensor): Query tensor (#batch, time1, size). key (torch.Tens

Compute encoded features. Args: x_input (torch.Tensor): Input tensor (#batch, time, size). mask (torch.Tensor): Mask

Embed positions in tensor.

Encoder + Decoder + Calc loss Args: speech: (Batch, Length, ...) speech_lengths: (Batch, ) tex

r""" Args: labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): Labels for computing

r""" labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for computing the sequence classification/regression

r""" labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for computing the sequence classification/regression

r""" start_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for position (index) of the start of the la

r""" Args: labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): Labels for computing

r""" labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for computing the sequence classification/regression

r""" labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for computing the sequence classification/regression

r""" start_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for position (index) of the start of the la

Compute scaled dot product attention. Args: query (torch.Tensor): Query tensor (#batch, time1, size). key (torch.Tens

Compute encoded features. Args: x_input (torch.Tensor): Input tensor (#batch, time, size). mask (torch.Tensor): Mask

Retrieve an item from the queue in a thread-safe manner. Parameters: - None Returns: - any or None: The ret

Returns the training [`~torch.utils.data.DataLoader`]. Will use no sampler if `train_dataset` does not implement `__len__`, a random

Check if the buffer contains enough data to fulfill a request of a specified length. Parameters: - length (int): The number