diff options
| author | Greg Heinrich <gheinrich@nvidia.com> | 2016-03-25 17:14:13 +0300 |
|---|---|---|
| committer | Greg Heinrich <gheinrich@nvidia.com> | 2016-03-25 23:00:24 +0300 |
| commit | 02bc67c446d0c42b7831a83ba15bff9677a372d2 (patch) | |
| tree | 64d91f02d33f955526e05a05815949826e53f72c /TemporalConvolution.lua | |
| parent | 7e74cb7464106114b468289f5dcc59744b2d25cd (diff) | |
Fix TemporalConvolution output size
Issue seen in updateOutput() when batch size is decreasing (e.g. 256->128):
```
.../torch/install/share/lua/5.1/torch/Tensor.lua:462: Wrong size for view. Input size: 66060288. Output size: 128x256x1008x1
stack traceback:
[C]: in function 'error'
.../torch/install/share/lua/5.1/torch/Tensor.lua:462: in function 'view'
...orch/install/share/lua/5.1/cudnn/TemporalConvolution.lua:62: in function <...orch/install/share/lua/5.1/cudnn/TemporalConvolution.lua:54>
```
See new test cudnntest.TemporalConvolution_reduceBatchSize() for a repro.
Calling a tensor's set() method without extra parameters but the requested storage causes the tensor to see a 1-D view of the full storage (not the associated tensor). For example:
```
th> x=torch.Tensor(10)
[0.0001s]
th> y=x:resize(5)
[0.0001s]
th> torch.Tensor():set(y:storage()):size()
10
[torch.LongStorage of size 1]
[0.0002s]
```
Proposed fix is to specify the desired view through the optional parameters of the set() method.
Diffstat (limited to 'TemporalConvolution.lua')
| -rw-r--r-- | TemporalConvolution.lua | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/TemporalConvolution.lua b/TemporalConvolution.lua index f59a5bd..a9e6470 100644 --- a/TemporalConvolution.lua +++ b/TemporalConvolution.lua @@ -57,7 +57,7 @@ function TemporalConvolution:updateOutput(input) self.buffer = self.buffer or torch.CudaTensor() self._output = self._output or torch.CudaTensor() if self.output:storage() then self._output:set(self.output:storage()) else self._output = self.output end - if self.buffer:storage() then self.output:set(self.buffer:storage()) else self.output = self.buffer end + if self.buffer:storage() then self.output:set(self.buffer:storage(), 1, self.output:size()) else self.output = self.buffer end cudnn.SpatialConvolution.updateOutput(self,_input) self.buffer = self.output:view(self.oSize):transpose(2,3) self.output = self._output:resize(self.buffer:size()):copy(self.buffer) |