Eoor while training for VG dataset

[anuragiitg]

I would like to train the model for VG dataset while training I am getting the following error.

Det: ../data/visualgenome/images/1029.jpg 565 800 [627 0 767 129] x2: 31 x1: 25 y2: 8 y1 :0 mask sum 48 value of (y2 - y1) * (x2 - x1) is 48 565 800 [ 86 358 564 934] x2: 23 x1: 3 y2: 32 y1 :20 mask sum 240 value of (y2 - y1) * (x2 - x1) is 240 565 800 [627 0 767 129] x2: 31 x1: 25 y2: 8 y1 :0 mask sum 48 value of (y2 - y1) * (x2 - x1) is 48 565 800 [ 0 676 173 1023] x2: 7 x1: 0 y2: 32 y1 :38 mask sum 0 value of (y2 - y1) * (x2 - x1) is -42 Traceback (most recent call last): File "train.py", line 120, in res.append((epoch,) + test_pre_net(net, args)+test_rel_net(net, args)) File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/model.py", line 112, in test_rel_net test_data = test_data_layer.forward() File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/data_layers/vrd_data_layer.py", line 56, in forward return self.forward_det() File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/data_layers/vrd_data_layer.py", line 192, in forward_det self._getDualMask(ih, iw, oBBox)]
File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/data_layers/vrd_data_layer.py", line 282, in _getDualMask assert(mask.sum() == (y2 - y1) * (x2 - x1)) AssertionError

Please provide the solution for the same. Thanking you in advance.

[Anakindai]

I would like to train the model for VG dataset while training I am getting the following error.

Det: ../data/visualgenome/images/1029.jpg 565 800 [627 0 767 129] x2: 31 x1: 25 y2: 8 y1 :0 mask sum 48 value of (y2 - y1) * (x2 - x1) is 48 565 800 [ 86 358 564 934] x2: 23 x1: 3 y2: 32 y1 :20 mask sum 240 value of (y2 - y1) * (x2 - x1) is 240 565 800 [627 0 767 129] x2: 31 x1: 25 y2: 8 y1 :0 mask sum 48 value of (y2 - y1) * (x2 - x1) is 48 565 800 [ 0 676 173 1023] x2: 7 x1: 0 y2: 32 y1 :38 mask sum 0 value of (y2 - y1) * (x2 - x1) is -42 Traceback (most recent call last): File "train.py", line 120, in res.append((epoch,) + test_pre_net(net, args)+test_rel_net(net, args)) File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/model.py", line 112, in test_rel_net test_data = test_data_layer.forward() File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/data_layers/vrd_data_layer.py", line 56, in forward return self.forward_det() File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/data_layers/vrd_data_layer.py", line 192, in forward_det self._getDualMask(ih, iw, oBBox)] File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/data_layers/vrd_data_layer.py", line 282, in _getDualMask assert(mask.sum() == (y2 - y1) * (x2 - x1)) AssertionError

Please provide the solution for the same. Thanking you in advance.

Have you solved it?

[anuragiitg]

Hi,

I have generated proposal file for VG to resolve the issue.

BR Anurag

On Sun, Nov 8, 2020 at 6:30 AM Anakindai [email protected] wrote:

I would like to train the model for VG dataset while training I am getting the following error.

Det: ../data/visualgenome/images/1029.jpg 565 800 [627 0 767 129] x2: 31 x1: 25 y2: 8 y1 :0 mask sum 48 value of (y2 - y1) * (x2 - x1) is 48 565 800 [ 86 358 564 934] x2: 23 x1: 3 y2: 32 y1 :20 mask sum 240 value of (y2 - y1) * (x2 - x1) is 240 565 800 [627 0 767 129] x2: 31 x1: 25 y2: 8 y1 :0 mask sum 48 value of (y2 - y1) * (x2 - x1) is 48 565 800 [ 0 676 173 1023] x2: 7 x1: 0 y2: 32 y1 :38 mask sum 0 value of (y2 - y1) * (x2 - x1) is -42 Traceback (most recent call last): File "train.py", line 120, in res.append((epoch,) + test_pre_net(net, args)+test_rel_net(net, args)) File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/model.py", line 112, in test_rel_net test_data = test_data_layer.forward() File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/data_layers/vrd_data_layer.py", line 56, in forward return self.forward_det() File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/data_layers/vrd_data_layer.py", line 192, in forward_det self._getDualMask(ih, iw, oBBox)] File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/data_layers/vrd_data_layer.py", line 282, in _getDualMask assert(mask.sum() == (y2 - y1) * (x2 - x1)) AssertionError

Please provide the solution for the same. Thanking you in advance.

Have you solved it?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/GriffinLiang/vrd-dsr/issues/38#issuecomment-723515617, or unsubscribe https://github.com/notifications/unsubscribe-auth/AC3IJADQT36GKC5KAJAI4J3SOXURNANCNFSM4G44DNNA .

[Anakindai]

Hi, I have generated proposal file for VG to resolve the issue. BR Anurag … On Sun, Nov 8, 2020 at 6:30 AM Anakindai @.***> wrote: I would like to train the model for VG dataset while training I am getting the following error. Det: ../data/visualgenome/images/1029.jpg 565 800 [627 0 767 129] x2: 31 x1: 25 y2: 8 y1 :0 mask sum 48 value of (y2 - y1) * (x2 - x1) is 48 565 800 [ 86 358 564 934] x2: 23 x1: 3 y2: 32 y1 :20 mask sum 240 value of (y2 - y1) * (x2 - x1) is 240 565 800 [627 0 767 129] x2: 31 x1: 25 y2: 8 y1 :0 mask sum 48 value of (y2 - y1) * (x2 - x1) is 48 565 800 [ 0 676 173 1023] x2: 7 x1: 0 y2: 32 y1 :38 mask sum 0 value of (y2 - y1) * (x2 - x1) is -42 Traceback (most recent call last): File "train.py", line 120, in res.append((epoch,) + test_pre_net(net, args)+test_rel_net(net, args)) File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/model.py", line 112, in test_rel_net test_data = test_data_layer.forward() File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/data_layers/vrd_data_layer.py", line 56, in forward return self.forward_det() File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/data_layers/vrd_data_layer.py", line 192, in forward_det self._getDualMask(ih, iw, oBBox)] File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/data_layers/vrd_data_layer.py", line 282, in _getDualMask assert(mask.sum() == (y2 - y1) * (x2 - x1)) AssertionError Please provide the solution for the same. Thanking you in advance. Have you solved it? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#38 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AC3IJADQT36GKC5KAJAI4J3SOXURNANCNFSM4G44DNNA .

I really appreciate for you kindness. By the way, could you share the script to genenrate .pkl or share the way to generate it? Thank you in advance!

[Anakindai]

I would like to train the model for VG dataset while training I am getting the following error.

Det: ../data/visualgenome/images/1029.jpg 565 800 [627 0 767 129] x2: 31 x1: 25 y2: 8 y1 :0 mask sum 48 value of (y2 - y1) * (x2 - x1) is 48 565 800 [ 86 358 564 934] x2: 23 x1: 3 y2: 32 y1 :20 mask sum 240 value of (y2 - y1) * (x2 - x1) is 240 565 800 [627 0 767 129] x2: 31 x1: 25 y2: 8 y1 :0 mask sum 48 value of (y2 - y1) * (x2 - x1) is 48 565 800 [ 0 676 173 1023] x2: 7 x1: 0 y2: 32 y1 :38 mask sum 0 value of (y2 - y1) * (x2 - x1) is -42 Traceback (most recent call last): File "train.py", line 120, in res.append((epoch,) + test_pre_net(net, args)+test_rel_net(net, args)) File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/model.py", line 112, in test_rel_net test_data = test_data_layer.forward() File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/data_layers/vrd_data_layer.py", line 56, in forward return self.forward_det() File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/data_layers/vrd_data_layer.py", line 192, in forward_det self._getDualMask(ih, iw, oBBox)] File "/home/siddharth/Documents/anurag/VRD/vrd-dsr/tools/../lib/data_layers/vrd_data_layer.py", line 282, in _getDualMask assert(mask.sum() == (y2 - y1) * (x2 - x1)) AssertionError

Please provide the solution for the same. Thanking you in advance.

I really appreciate for you kindness. By the way, could you share the script to genenrate .pkl or share the way to generate it? Thank you in advance!