Markus Kängsepp

Hey Niklas, The reliability diagrams for this work are generated using the following code: https://github.com/markus93/NN_calibration/blob/master/scripts/calibration/Reliability%20Diagram.ipynb. There are some other examples available at [here](https://github.com/dirichletcal/experiments_neurips/blob/master/notebooks/toy_example_experiment.ipynb ) (cell 25), using method "_plot_multiclass_reliability_diagram_", which...

Hey, sorry for the late response. I am not sure if it is the issue, but the pickled logits file ((logits_val, y_val), (logits_test, y_test)) should contain the following: - `logits_val`...

Hey Mariana, Thanks for the question, there is also an alternative way of downloading the logits from Google Drive https://drive.google.com/drive/folders/1e1ai-bKb7LukKShqpn3S_gYXJGzhUYgm?usp=sharing.

Hey, sorry for a late answer. In the script [https://github.com/markus93/NN_calibration/blob/master/scripts/utility/evaluation.py](url) is a method "evaluate_model". You can use this to get the logits out from the model and it also saves...

Hey! Most of the models are available [here ](https://owncloud.ut.ee/owncloud/s/8wjqyeM2aKfZoGp). I will add this link also to the models' folder. Best wishes, Markus

Hey! This is quite old code. Thus, it uses an older version of Keras (likely Keras 2.3.1), and other packages could influence the behaviour. The problem is that lines 30-34...

You could try model._layers.pop() instead of model.layers.pop(), maybe that does the trick. Or some other workaround from here https://stackoverflow.com/questions/57316557/tf-keras-layers-pop-doesnt-work-but-tf-keras-layers-pop-does.