How to get a matrix of two dynamic axes
Feature request: NumPy Reader
Cntk can not detect my gpu by cntk.all_devices()
Why is CNTK using the embedding dimension for the decoder?
How to use gather on 1D vector?
How to get the shape / dimension of the layers?
For CNTK sequence to sequence model, what is the samples per second meaning?
access trained parameter in CNTK
The same matrix with dim [0, 0] has been transferred between different devices for 20 times
How can I save a CNTK model as text?
CNTK: Applying average pooling over LSTM output
How to do minibatch evaluation using CNTK eval api
CNTK always return errorRate is 0
How to use Sequential to without weight sharing in Layers Library?
Dynamic Axes with a custom RNN
WARNING: Exhausted all input expected for the current sequence while reading a floating point value at offset