EmilPi

Apollo optimizer eats all the GPU memory

2

My network is: a few dense layers (conv with padding + concatenating output to input), 2-layer LSTM and 2 Linear layers in the end. Even after I made a network...

I install PuzzleLib from within PuzzleLib path with command `pip3 install ./ --install-option="--backend=cuda"` . Installation goes without error. Now, when I run `python3 TestLib/CnnCifar10NIN.py`, I get Traceback (most recent call...

Can't compile with CUDA 10.0

2

Trying to compile https://github.com/che-shr-cat/currennt/ with CUDA 10.0: ``` git clone https://github.com/che-shr-cat/currennt/ mkdir build && cd build && cmake -DCUDA_cublas_device_LIBRARY=/usr/local/cuda/targets/x86_64-linux/lib/libcublas.so .. && make ``` yields ``` ... [ 61%] Building NVCC...

It is warning in CUDA 10.0, but it may fail to compile with CUDA

I am interested in supporting RNNs in C++ without need to use DL frameworks. Maybe let's tweak this repo a bit :)