Extend support to Python3 and PyTorch 0.4.1

[SergioRAgostinho]

Hi,

I just made some modifications in order to be able to run valid.py and train.py using Python3 and the most recent versions of PyTorch. I believe all changes proposed are backwards compatible with the original targeted versions of Python and PyTorch, mentioned in README.md.

With the changes proposed, running

python3 valid.py cfg/ape.data cfg/yolo-pose.cfg backup/ape/model_backup.weights

with the original data provided, produced the following output.

layer     filters    size              input                output
    0 conv     32  3 x 3 / 1   416 x 416 x   3   ->   416 x 416 x  32
    1 max          2 x 2 / 2   416 x 416 x  32   ->   208 x 208 x  32
    2 conv     64  3 x 3 / 1   208 x 208 x  32   ->   208 x 208 x  64
    3 max          2 x 2 / 2   208 x 208 x  64   ->   104 x 104 x  64
    4 conv    128  3 x 3 / 1   104 x 104 x  64   ->   104 x 104 x 128
    5 conv     64  1 x 1 / 1   104 x 104 x 128   ->   104 x 104 x  64
    6 conv    128  3 x 3 / 1   104 x 104 x  64   ->   104 x 104 x 128
    7 max          2 x 2 / 2   104 x 104 x 128   ->    52 x  52 x 128
    8 conv    256  3 x 3 / 1    52 x  52 x 128   ->    52 x  52 x 256
    9 conv    128  1 x 1 / 1    52 x  52 x 256   ->    52 x  52 x 128
   10 conv    256  3 x 3 / 1    52 x  52 x 128   ->    52 x  52 x 256
   11 max          2 x 2 / 2    52 x  52 x 256   ->    26 x  26 x 256
   12 conv    512  3 x 3 / 1    26 x  26 x 256   ->    26 x  26 x 512
   13 conv    256  1 x 1 / 1    26 x  26 x 512   ->    26 x  26 x 256
   14 conv    512  3 x 3 / 1    26 x  26 x 256   ->    26 x  26 x 512
   15 conv    256  1 x 1 / 1    26 x  26 x 512   ->    26 x  26 x 256
   16 conv    512  3 x 3 / 1    26 x  26 x 256   ->    26 x  26 x 512
   17 max          2 x 2 / 2    26 x  26 x 512   ->    13 x  13 x 512
   18 conv   1024  3 x 3 / 1    13 x  13 x 512   ->    13 x  13 x1024
   19 conv    512  1 x 1 / 1    13 x  13 x1024   ->    13 x  13 x 512
   20 conv   1024  3 x 3 / 1    13 x  13 x 512   ->    13 x  13 x1024
   21 conv    512  1 x 1 / 1    13 x  13 x1024   ->    13 x  13 x 512
   22 conv   1024  3 x 3 / 1    13 x  13 x 512   ->    13 x  13 x1024
   23 conv   1024  3 x 3 / 1    13 x  13 x1024   ->    13 x  13 x1024
   24 conv   1024  3 x 3 / 1    13 x  13 x1024   ->    13 x  13 x1024
   25 route  16
   26 conv     64  1 x 1 / 1    26 x  26 x 512   ->    26 x  26 x  64
   27 reorg              / 2    26 x  26 x  64   ->    13 x  13 x 256
   28 route  27 24
   29 conv   1024  3 x 3 / 1    13 x  13 x1280   ->    13 x  13 x1024
   30 conv     20  1 x 1 / 1    13 x  13 x1024   ->    13 x  13 x  20
   31 detection
2018-08-06 15:33:48    Testing ape...
2018-08-06 15:33:48    Number of test samples: 1050
-----------------------------------
  tensor to cuda : 0.000332
         predict : 0.002955
get_region_boxes : 0.051849
            eval : 0.014257
           total : 0.069392
-----------------------------------
2018-08-06 15:35:02 Results of ape
2018-08-06 15:35:02    Acc using 5 px 2D Projection = 94.48%
2018-08-06 15:35:02    Acc using 10% threshold - 0.0103 vx 3D Transformation = 28.00%
2018-08-06 15:35:02    Acc using 5 cm 5 degree metric = 51.90%
2018-08-06 15:35:02    Mean 2D pixel error is 2.833865, Mean vertex error is 0.029564, mean corner error is 4.058189
2018-08-06 15:35:02    Translation error: 0.029320 m, angle error: 5.689846 degree, pixel error:  2.833865 pix

I looks ok-ish, but I would be good to have some confirmation from your side that was no deterioration in performance. I will be performing the porting to the training routine later on.

[msftclas]

All CLA requirements met.

[juanmed]

Making these changes for the multi object pose estimation task seem to work as well.

[CVUsers]

@SergioRAgostinho

Hello, how to use it on the latest Python?

In addition, I always report this error: output = model (data). Data: runtimeerror: cudnn error: cudnn_ STATUS_ MAPPING_ ERROR