I've trained an MNIST network. Now I basically want to know how I can input an image and classify it. By reading the docs, I understand that provide_external_data is one of the steps of doing that.
However, I find think that the documentation does not explain clearly enough the data parameter that net.provide_external_data takes in. I understand that it should be a dictionary, but I'm not sure what the dictionary should contain. An quick example dictionary (f.ex. for the MNIST problem) and a reference to the data format documentation would help.
I've trained an MNIST network. Now I basically want to know how I can input an image and classify it. By reading the docs, I understand that provide_external_data is one of the steps of doing that.
However, I find think that the documentation does not explain clearly enough the data parameter that net.provide_external_data takes in. I understand that it should be a dictionary, but I'm not sure what the dictionary should contain. An quick example dictionary (f.ex. for the MNIST problem) and a reference to the data format documentation would help.