Adding custom features to clstm
bluefa1con opened this issue · 0 comments
bluefa1con commented
I am training clstm over the IAM handwriting database and need to add more features like:
The mean grey value of the pixels
The centre of gravity of the pixels
The second order vertical moment of the centre of gravity
The positions of the uppermost and lowermost black
pixels
The rate of change of these positions (with respect to the
neighbouring windows)
These features are to be calculated at every position of the sliding window
How would I get around to doing that? Should I add them to the same tensor clstm uses to store pixel values and increase its dimension or introduce more tensors and if so where would i go around to passing them to the network
@tmbdev can you help with this please