fl16180 / small-object-detection Goto Github PK

F:\softwares\Anaconda3\envs\py36\lib\site-packages\torch\nn_reduction.py:43: UserWarning: size_average and reduce args will be deprecated, please use reduction='none' instead.
warnings.warn(warning.format(ret))
Traceback (most recent call last):
File "train.py", line 58, in
main(config)
File "train.py", line 39, in main
trainer.train()
File "D:\GitHub\my new code\base\base_trainer.py", line 65, in train
result = self.train_epoch(epoch)
File "D:\GitHub\my new code\trainer\trainer.py", line 54, in train_epoch
output_boxes, output_scores = self.model(data)
File "F:\softwares\Anaconda3\envs\py36\lib\site-packages\torch\nn\modules\module.py", line 532, in __ call
result = self.forward(*input, **kwargs)
File "D:\GitHub\my new code\model\model.py", line 450, in forward
spatial_corr4 = self.spatialcooc4(conv4_out)
File "F:\softwares\Anaconda3\envs\py36\lib\site-packages\torch\nn\modules\module.py", line 532, in call
result = self.forward(*input, **kwargs)
File "D:\GitHub\my new code\model\cooc_layers.py", line 101, in forward
channel_pairs = channel_pairs.view(-1,2)
RuntimeError: view size is not compatible with input tensor's size and stride (at least one dimension spans across two contiguous subspaces). Use .reshape(...) instead.

RuntimeError: Error(s) in loading state_dict for SSD300:
Unexpected key(s) in state_dict: "spatialcooc4.conv1.weight", "spatialcooc4.gaussian.weight", "spatialcooc7.conv1.weight", "spatialcooc7.gaussian.weight", "spatialcooc8.conv1.weight", "spatialcooc8.gaussian.weight".
size mismatch for classifiers.box4.weight: copying a param with shape torch.Size([16, 576, 3, 3]) from checkpoint, the shape in current model is torch.Size([16, 512, 3, 3]).
size mismatch for classifiers.box7.weight: copying a param with shape torch.Size([24, 1088, 3, 3]) from checkpoint, the shape in current model is torch.Size([24, 1024, 3, 3]).
size mismatch for classifiers.box8.weight: copying a param with shape torch.Size([24, 576, 3, 3]) from checkpoint, the shape in current model is torch.Size([24, 512, 3, 3]).
size mismatch for classifiers.class4.weight: copying a param with shape torch.Size([8, 576, 3, 3]) from checkpoint, the shape in current model is torch.Size([8, 512, 3, 3]).
size mismatch for classifiers.class7.weight: copying a param with shape torch.Size([12, 1088, 3, 3]) from checkpoint, the shape in current model is torch.Size([12, 1024, 3, 3]).
size mismatch for classifiers.class8.weight: copying a param with shape torch.Size([12, 576, 3, 3]) from checkpoint, the shape in current model is torch.Size([12, 512, 3, 3]).