diff options
| author | Anthony Sandrin <asandrin@nvidia.com> | 2016-03-12 02:29:49 +0300 |
|---|---|---|
| committer | Boris Fomitchev <bfomitchev@nvidia.com> | 2016-04-18 23:18:26 +0300 |
| commit | 15cec07968175f2edc5368bca32829319ad0570b (patch) | |
| tree | a3dc910d9c7d22932daa6e5c04bb77a41397db7a /test | |
| parent | 19dc9a13b5812cd74016a3b9e27d306f35b4b42f (diff) | |
Initial work for cudnn RNN api integration
Added cudnnFind auto-tuning
Change RNN layer api and improve descriptor/tensor resizing conditions
Implement updateGradInput and accGradParameters
Diffstat (limited to 'test')
| -rw-r--r-- | test/test.lua | 5 |
1 files changed, 3 insertions, 2 deletions
diff --git a/test/test.lua b/test/test.lua index 9b5cbde..8fcd1b9 100644 --- a/test/test.lua +++ b/test/test.lua @@ -1444,9 +1444,10 @@ math.randomseed(os.time()) mytester = torch.Tester() mytester:add(cudnntest) -if torch.random(1,2) == 1 then +-- if torch.random(1,2) == 1 then cudnn.benchmark = true -- run manual auto-tuner -end + cudnn.verbose = true +-- end for i=1,cutorch.getDeviceCount() do |