I ran the meshCNN test and got RuntimeError: Invalid index in gather.
The training data was self-produced and tested after learning, but an error message occurred. Can you possibly tell the cause?
(MeshCNN) C:\project_workspace\MeshCNN-master>test_dental.cmd run the test and export collapses Running Test base_init loaded mean / std from cache loading the model from ./checkpoints\dental_seg\latest_net.pth Traceback (most recent call last): File "test.py", line 25, in <module> run_test() File "test.py", line 18, in run_test ncorrect, nexamples = model.test() File "C:\project_workspace\MeshCNN-master\models\mesh_classifier.py", line 116, in test correct = self.get_accuracy(pred_class, label_class) File "C:\project_workspace\MeshCNN-master\models\mesh_classifier.py", line 125, in get_accuracy correct = seg_accuracy(pred, self.soft_label, self.mesh) File "C:\project_workspace\MeshCNN-master\util\util.py", line 28, in seg_accuracy correct_mat = ssegs.gather(2, predicted.cpu().unsqueeze(dim=2)) RuntimeError: Invalid index in gather at C:\w\1\s\tmp_conda_3.6_045031\conda\conda-bld\pytorch_1565412750030\work\aten\src\TH/generic/THTensorEvenMoreMath.cpp:472
I ran the meshCNN test and got RuntimeError: Invalid index in gather.
The training data was self-produced and tested after learning, but an error message occurred. Can you possibly tell the cause?
(MeshCNN) C:\project_workspace\MeshCNN-master>test_dental.cmd run the test and export collapses Running Test base_init loaded mean / std from cache loading the model from ./checkpoints\dental_seg\latest_net.pth Traceback (most recent call last): File "test.py", line 25, in <module> run_test() File "test.py", line 18, in run_test ncorrect, nexamples = model.test() File "C:\project_workspace\MeshCNN-master\models\mesh_classifier.py", line 116, in test correct = self.get_accuracy(pred_class, label_class) File "C:\project_workspace\MeshCNN-master\models\mesh_classifier.py", line 125, in get_accuracy correct = seg_accuracy(pred, self.soft_label, self.mesh) File "C:\project_workspace\MeshCNN-master\util\util.py", line 28, in seg_accuracy correct_mat = ssegs.gather(2, predicted.cpu().unsqueeze(dim=2)) RuntimeError: Invalid index in gather at C:\w\1\s\tmp_conda_3.6_045031\conda\conda-bld\pytorch_1565412750030\work\aten\src\TH/generic/THTensorEvenMoreMath.cpp:472
Option setting python test.py --dataroot datasets/dental_seg --name dental_seg --arch meshunet --dataset_mode segmentation --ncf 32 64 128 256 --ninput_edges 5714 --pool_res 5000 4000 2000 --resblocks 3 --batch_size 1 --export_folder meshes