No need of big input.

Allow fallback on GpuCorrMM if cuDNN is not there

Use FAST_RUN for reference version.

Fix parameter types in ave_pool kernels, remove static_cast, add GLOBAL_MEM

This should speed up the FAST_COMPILE buildbot

Fail if output memory not allocated

Fix elemwise ValueError message (printf formatting)

This fixes a compilation issue on Pascal GPUs.

instead of segfaulting like a barbarian.

Update documentation and config flags about supported cuDNN algorithms.

Don't print float16 warning for ops that don't have c code

jenkins buildbot with docker

Add lifter for CrossentropyCategorical1Hot and grad

Added mode 'half' to Images2Neibs. Tests pass.

[ENH] Speed up nanguardmode by not checking *int* dtype

(small fix) Add doc for `dnn.include_path` and `dnn.library_path`.

Split long-running test so that it helps travis not to give up

Fix offset problems in the new backend.

Fix issue in optimizations with bitwise operations

Port Softmax kernel to OpenCL

All tests in test_nnet.py pass with CUDA.
Only fp32 tests in test_nnet.py pass with OpenCL. GpuFromHost doesn't work with fp16 or fp64.
Larger work item size doesn't improve performance.
Add 2 local_barrier(), it's strange that AMD card doesn't need these local_barrier(), but they are necessary for NVIDIA cards.