rn-4.2.4

Merge remote-tracking branch 'zakki/nn' into nn

rn-4.2.1

Ray 8.0 with DCNN
The binary is for 64 bits Windows with NVIDIA GPU.
Rn runs without GPU, but it is recommended to use GTX 970 or later.

Microsoft Visual C++ 2015 Redistributable is required.

Use --thread <N> --reuse-subtree to use multi-core CPU.
--ponder is not tested well.

rn-3.11.1

Ray 8.0 with DCNN
The binary is for 64 bits Windows with NVIDIA GPU.
Rn runs without GPU, but it is recommended to use GTX 970 or later.

Microsoft Visual C++ 2015 Redistributable is required.

Use --thread <N> --reuse-subtree to use multi-core CPU.
--ponder is not tested well.

Rn.3.10.2

Ray 8.0 with DCNN
Binary is for 64 bits Windows with nvidia GPU.

Microsoft Visual C++ 2015 Redistributable is required.

Use --thread <N> --reuse-subtree to use multi-core CPU.
--ponder is not tested well.

rn-3.3

If you don't use a cuda capable card, update dlls with CNTK 2.0. beta 9.0. to fix batch normalization. https://github.com/Microsoft/CNTK/wiki/CNTK_2_0_beta_8_Release_Notes

rayon 2.3

Ray 8.0 with DCNN