libdnn
libdnn copied to clipboard
poweic
A lightweight and user friendly C++ library for deep and convolutional neural network with GPU acceleration
install-sh failed on an incorrect option, ctags didn't support --langmap: ``` ctags: unrecognized option '--langmap=C:+.cu' ``` This caused make to fail, not revert to the project root directory and download...
I noticed you wanted to switch to cuDNN. I would like to keep the old CUDA code as well. For example my GPU doesn't allow for cuDNN (it's too old)....
I was wondering why you added 1 to the AffineTransform layer geometry in https://github.com/botonchou/libdnn/blob/master/src/nnet.cpp#L233 After some nosing around I found you feed the label into the network as being one...
I had followed all steps in FAQ and Quick Start, but I got the message: no rule to make target 'o3', Stop Other simple question: What path did you install...
In `class NNet`, I use `std::vector` to store all the feature transformations. For Recurrent Neural Network (RNN) or other user-defined NN structures, `std::vector` is not enough. A graph container and...
Hi Chou, It seems that the libdnn couldn't handle the multi-label task. Would you like to add this new feature in the future?
Consider adding input-dim and output-dim to the --struct parameter. So: `--input-dim 4 --output-dim 2 10-10` Becomes: `4-10-10-2` In my opinion this makes for a cleaner syntax. Why are they separate...
Gains in the last affine layer seem to drop to +- minus 50. https://gist.github.com/supergrover/733504ee836506603ea4
I don't see that using Apache license would do the library, as a grassroot project, any good in general. First of all, Apache license as far as I know it...