GitLab

clean up PT <=> TF conversion

[WIP] Squad refactor

create encoder attention mask from shape of hidden states

We currently save the pretrained_weights of the encoder and decoder in
two separate directories `encoder` and `decoder`. However, for the
`from_pretrained` function to operate with automodels we need to
specify the type of model in the path to the weights.

The path to the encoder/decoder weights is handled by the
`PreTrainedEncoderDecoder` class in the `save_pretrained` function. Sice
there is no easy way to infer the type of model that was initialized for
the encoder and decoder we add a parameter `model_type` to the function.
This is not an ideal solution as it is error prone, and the model type
should be carried by the Model classes somehow.

This is a temporary fix that should be changed before merging.

Since I started my PR the `add_special_token_single_sequence` function
has been deprecated for another; I replaced it with the new function.

We currently create encoder attention masks (when they're not provided)
based on the shape of the inputs to the encoder. This is obviously
wrong; sequences can be of different lengths. We now create the encoder
attention mask based on the batch_size and sequence_length of the
encoder hidden states.