Robin van Schaik

Any updates on this / where it is on the roadmap? I agree, this tool is excellent and the correlations are the only thing missing at the moment. As such,...

Hi @sgoede, How nice to hear. Thanks for adding a comment to illustrate this issue. Are you willing to contribute your solution with a PR? :)

Hi @cpuyyp, Thanks for using the repo and raising the issue. I may have overlooked it because my Macbook does not have a GPU. I will probably adjust it this...

Hi @cpuyyp, I tested that this works indeed on a GPU notebook on Google Cloud. ```Python # Store the encoding on the GPU. input_ids = input_ids.to(flair_model_wrapper.device) ```  Not sure...

> In my case your explanation helped (I had to edit the function manually as you proposed after cloning your repo, so I guess this solution is not in the...

Hi @krzysztoffiok, Thanks for contributing with your pull requests! Keep them coming. Are you willing to reflect your changes in the tutorial in the readme.md as well? You might have...

@krzysztoffiok You are definitely right. I should have added a requirements.txt to the repository with pinned versions. This would make it more reproducible.

With a [1GB cap](https://docs.github.com/en/free-pro-team@latest/github/managing-large-files/about-storage-and-bandwidth-usage) this does not really seem like a suitable solution.

Obviously, Flair also has a built-in sentiment classifier model. I will rewrite the readme to use this model.