Atif Ahmed

Adding BERT for MS-MARCO Passage re-ranking

2

Adding BERT fine-tuned on MS-MARCO for passage re-ranking task (https://arxiv.org/abs/1901.04085) Since this is a pretrained classifier, we had to add final linear layer parameters in `PretrainedBERTMixin`. Based on the `pretrained_model_name`,...

There are many processors which are testing very similar things. E.g. all the tokenizer tests are testing the tokens created by those processors which are same for our test cases....

There are many processors which are doing very similar things. E.g. some of the 3rd party tokenizers. Even though the 3rd party interfaces might differ, there still can be code...