Type error in compositionality

[prata-uece]

The try-except was removed from the following block: try: succ_grad = evaluate_derivatives(pathdir,filename,model_feynman) except: succ_grad = 0 Execution error at grad_evaluated = evaluate_derivatives(directory, filename, model_feynman) File “.../site-packages/torch/nn/functional.py”, line 1847, in linear return torch._C._nn.linear(input, weight, bias) RuntimeError: expected scalar type Float but found Double Operation: evaluate_derivatives (in S_NN_get_gradients.py), after printing “Checking for compositionality...” Line: pts = torch.tensor(pts) Partial traceback: File “.../site-packages/torch/nn/functional.py”, line 1847, in linear return torch._C._nn.linear(input, weight, bias) RuntimeError: expected scalar type Float but found Double The above error occurs every time “evaluate_derivatives” is called, which implies that “compositionality” never succeeds. Request: would it be possible, if this bug is corrected, to also update the code on PyPI?