The TensorFlow implementation of the LXMERT model currently has no integration tests. This is problematic as the behavior can diverge without being noticed.

The test_modeling_tf_lxmert.py file should be updated to include integration testing.

An example of a good modeling integration test is visible i

From paper, it mentioned

Instead, the training data generator chooses 15% of tokens at random, e.g., in the sentence my
dog is hairy it chooses hairy.

It means that 15% of token will be choose for sure.

From https://github.com/codertimo/BERT-pytorch/blob/master/bert_pytorch/dataset/dataset.py#L68,
for every single token, it has 15% of chance that go though the followup procedure.

Is your feature request related to a problem? Please describe.

One of the typical use cases that we saw in the community is using Haystack to boost conversational assistants / chat bots on the long tail of queries. As you can't think of all possible intents beforehand, a QA model is a powerful option to cover unforeseen "information queries".

The integration of Haystack with the existin

Issue to track tutorial requests:

• Deep Learning with PyTorch: A 60 Minute Blitz - #69