DIGITS Train Caffe model fail, CUDNN_STATUS_BAD_PARAM (resolved)

Accelerated Computing GPU-Accelerated Libraries
1

…

2

If you installed cuDNN properly, it should be using cuDNN either with LeNet or with AlexNet.

This concerns me:

"I wanted to try the same dataset on the AlexNet, "

The lenet configuration is a 28x28 input, which is appropriate for a 32x32 mnist training set. If you try to use the same dataset with the AlexNet model, I think you’re going to have trouble. Isn’t the AlexNet model set up for a 256x256 data set?

3

…