RuntimeError: cannot perform reduction function argmax on a tensor with no elements because the operation does not have an identity
1s23 opened this issue · comments
`Error executing job with overrides: ['+experiments=dd3d_kitti_dla34_overfit', 'EVAL_ONLY=True', 'MODEL.CKPT=/home/cgim/yyh/github/dd3d/model_final.pth', 'TEST.IMS_PER_BATCH=4']
Traceback (most recent call last):
File "./scripts/train.py", line 60, in main
test_results = do_test(cfg, model, is_last=True)
File "./scripts/train.py", line 232, in do_test
per_dataset_results = inference_on_dataset(model, dataloader, evaluator)
File "/home/cgim/anaconda3/envs/yyh/lib/python3.6/site-packages/detectron2/evaluation/evaluator.py", line 158, in inference_on_dataset
outputs = model(inputs)
File "/home/cgim/anaconda3/envs/yyh/lib/python3.6/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/cgim/yyh/github/dd3d/tridet/modeling/dd3d/core.py", line 121, in forward
fcos2d_info
File "/home/cgim/yyh/github/dd3d/tridet/modeling/dd3d/fcos3d.py", line 327, in call
pred_instances[lvl], fcos2d_info[lvl]
File "/home/cgim/yyh/github/dd3d/tridet/modeling/dd3d/fcos3d.py", line 408, in forward_for_single_feature_map
depth_is_distance=self.predict_distance
File "/home/cgim/yyh/github/dd3d/tridet/modeling/dd3d/fcos3d.py", line 48, in predictions_to_boxes3d
quat = allocentric_to_egocentric(quat, proj_ctr, inv_intrinsics)
File "/home/cgim/yyh/github/dd3d/tridet/utils/geometry.py", line 47, in allocentric_to_egocentric
egocentric_quat = matrix_to_quaternion(R_obj_to_global)
File "/home/cgim/anaconda3/envs/yyh/lib/python3.6/site-packages/pytorch3d/transforms/rotation_conversions.py", line 151, in matrix_to_quaternion
F.one_hot(q_abs.argmax(dim=-1), num_classes=4) > 0.5, : # pyre-ignore[16]
RuntimeError: cannot perform reduction function argmax on a tensor with no elements because the operation does not have an identity
Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.`
Can you tell me how to solve this problem? When I use this model I get an error like this
@1s23 hi, I met the same problem, Do you have solved it now?