Discriminative Attribution Dataset B

This is a synthetic dataset built to benchmark the capabilities of explainable AI methods against difficult tasks.<br>The dataset has three classes, which depend absence of a given shape (circle, square, or triangle) in the image.<br>Also attached are the PyTorch model weights of a classifier trained on the task, both the raw weights and a JIT-compiled version. The raw weights can be loaded with the Funke Lab torch package.