SJ001/AI-Feynman

final_gd signature is wrong

Closed this issue · 0 comments

final_gd takes data as argument:

def final_gd(data, math_expr, lr = 1e-2, N_epochs = 5000):

but it is called with the pathdir and filename:

gd_update = final_gd(pathdir,filename,PA_list[i][-1])

I imagine this is not caught due to the exception handling, but that also means the final gradient descent never happens.