Problem running MNIST classifier example using Keras and TensorFlow extensions

[ienet1308]

Dear All,

I am having trouble running the following KNIME deep learning example 04_Analytics\14_Deep_Learning\02_Keras\03_Train_MNIST_classifier from the example server.

The error message is [Selected input columns do not provide enough element(10) to populate all neurons (784) of network input ‘conv2d_1_input(0).’ Try adding some columns to the selection.]

I using KNIME 3.6.0, python 3.5.5, tensorflow 1.8.0 and keras 2.1.6. Any idea how to solve it?

Thanks in advance,
Regards.

[ienet1308]

I find the answer by myseIf. The root cause is that I didn’t install knime image processing extension.