Transformers Text Classification Example: Compute Precision, Recall and F1


I am using this text classification tutorial from Huggingface with my own train, test and validation data.

My goal is to compute a classification report with all the metrics in it (F1, accuracy, Recall, Precision), but I don’t know how to do this with this example script.

The script runs well and at the end i get following message: “INFO - main - Wrote predictions to output/test_results.txt!”

In the “output”-Folder I see these files: “config.json”, “test_results.txt” and “tf_model.h5”.

So how can I can evaluate my trained model?

Thanks in advance!!