Julian Eisenschlos

Hello @voldermort28 the script accepts `--since` and `--until` parameters to limit the dates to get messages, but currently it doesn't support a limit on quantity. We could add a paramenter...

Hi, I think there are multiple ways of doing what you want. One way would be to manipulate the checkpoint to change the shape of the `output_agg_bias` and `output_weights_agg` adding...

I am not sure sadly, consider asking in the tensorflow repo. Perhaps [this](https://stackoverflow.com/questions/48138041/modifying-shape-of-tensor-in-tensorflow-checkpoint) can be of help.

Hello @NielsRogge if you want to add more information to the model output you can do so in the `predictions` map in [tapas_classifier_model.py](https://github.com/google-research/tapas/blob/083b26d7f60204d41738fb3f280a958f986d2080/tapas/models/tapas_classifier_model.py#L1034) for example by adding `model.pooled_output()`, `model.sequence_output()` or...

Hello @jaihonikhil regarding the first issue, as you may imagine the problem is that changing the number of classes requires changing the dimension of the last projection, thus producing an...

Hi @mohandas1 can you explain in more detail what's the issue, ideally with a colab to reproduce the error. A model that computes aggregation functions, such as the one trained...

Hi @joshplasse can you share some information about your dataset, is it public? How big is it? By altering the number of rows and length solely for fine-tuning some of...

Thanks for the interest! We are working on it but nothing ready at the moment. Stay tuned, will ping this issue once it is released.

Hi there, we don't have such statistics at the moment. For the paper's experiments, we used a brute-force nearest neighbour search, which in a production application can be dramatically improved...

Thanks for your interest. Exporting a saved model to tensorflow.js is certainly the most direct way, but that said we haven't tried this so far. I would be very interested...