Ensure attention mask is assigned to GPU

Ensure consistency of device assignment when training

Reduced batch size

Created HuggingFace Space

Ensure consitency value is 32 bit

Create BatchEncoding in pad

Use BatchEncoding for training, not batch_encoding

Use BatchEncoding for training

Removed unnecessary on_epoch_end

Removed superfluous ()

Modified CombinedCorpus to use PyTorch

items, not values

Use dictionary coprehensions to do padding and return dictionaries instead of default dictionaries

Implement max_lengths for CorpusRepeater

Forgot a len

TPUs need constant batch shapes

Incoplete testing script for consistency. Fixed typo

Attention masks, generation, and testing script

Attention masks are only necessary for inputs

Making sure RoBERTa layers have all required arguments

Removed unnecessary ()

Forgot ()

Another workaround

itertools.batched not available

Lazy batching to prevent memory errors

Fixed indentation error

Encoding.pad modifies in place

Fix typo

columns is a list

columns is a local variable

Don't forget the label

Fix typo

Fix typo

Fix typo

Tokenization gotcha

Missing for clause in dict comprehension

Don't forget to import pandas

Fix typo

Fix imports

Decided to use pandas rather than datasets

Components for managing training corpora

CorpusRepeater

Code for building and training base modkels.

Code for building and training base modkels.