In two_layer_net.ipynb , when I use a Solver instance to train a TwoLayerNet, a numerical overflow error raised, which caused the softmax loss nan, but I have successfully passed all test above this step. Anyone also encountered this bug? Please give me some advice!
In
two_layer_net.ipynb
, when I use a Solver instance to train a TwoLayerNet, a numerical overflow error raised, which caused the softmax lossnan
, but I have successfully passed all test above this step. Anyone also encountered this bug? Please give me some advice!