search

lufficc / SSD

High quality, fast, modular reference implementation of SSD in PyTorch
MIT License
1.52k stars 382 forks source link

IndexError: The shape of the mask [12, 1] at index 1 does not match the shape of the indexed tensor [12, 24564] at index 1 #201

Open weimengmeng1999 opened 3 years ago

weimengmeng1999 commented 3 years ago

Hi, I'm using my own dataset to train the model. But I got the following error when during training, which I have no idea what's wrong. I've checked all the datatype, and they are all the same as the ones in VOC and COCO. Could you please help me with it:

Traceback (most recent call last): File "train.py", line 114, in main() File "train.py", line 105, in main model = train(cfg, args) File "train.py", line 44, in train model = do_train(cfg, model, train_loader, optimizer, scheduler, checkpointer, device, arguments, args) File "/home/tribta/SSD/ssd/engine/trainer.py", line 82, in do_train loss_dict = model(images, targets=targets) File "/home/tribta/miniconda3/envs/wmod/lib/python3.8/site-packages/torch/nn/modules/module.py", line 532, in call result = self.forward(*input, kwargs) File "/home/tribta/SSD/ssd/modeling/detector/ssd_detector.py", line 16, in forward detections, detector_losses = self.box_head(features, targets) File "/home/tribta/miniconda3/envs/wmod/lib/python3.8/site-packages/torch/nn/modules/module.py", line 532, in call result = self.forward(*input, *kwargs) File "/home/tribta/SSD/ssd/modeling/box_head/box_head.py", line 25, in forward return self._forward_train(cls_logits, bbox_pred, targets) File "/home/tribta/SSD/ssd/modeling/box_head/box_head.py", line 31, in _forward_train reg_loss, cls_loss = self.loss_evaluator(cls_logits, bbox_pred, gt_labels, gt_boxes) File "/home/tribta/miniconda3/envs/wmod/lib/python3.8/site-packages/torch/nn/modules/module.py", line 532, in call result = self.forward(input, kwargs) File "/home/tribta/SSD/ssd/modeling/box_head/loss.py", line 32, in forward mask = box_utils.hard_negative_mining(loss, labels, self.neg_pos_ratio) File "/home/tribta/SSD/ssd/utils/box_utils.py", line 122, in hard_negative_mining loss[pos_mask] = -math.inf IndexError: The shape of the mask [12, 1] at index 1 does not match the shape of the indexed tensor [12, 24564] at index 1

palealice commented 2 years ago

me too

yongsenli commented 1 year ago

me too,so,how can i do for this?