Check if you have training samples in your test set
Did you manage to spill samples from your train set to your test set?
from did_it_spill import check_spill
spills = check_spill(train_loader, test_loader)
print(f"You have {len(spills)} spills in your test set!")
The library computes hashes of your data to determine if you have samples spilled over from your train set to test set.
Currently only for PyTorch.
Installation
Outputs
Function outputs a list of tuples. Each tuple corresponds to a leak. The first index is where in the first loader the
leak was found, and the second index is the index