Start training...
Starting epoch 0
/home/mmc_xiaolinhui/mmc_226_exp_202206/mdetr/models/position_encoding.py:41: UserWarning: floordiv is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
dim_t = self.temperature ** (2 * (dim_t // 2) / self.num_pos_feats)
/home/mmc_xiaolinhui/mmc_226_exp_202206/mdetr/models/position_encoding.py:41: UserWarning: floordiv is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
dim_t = self.temperature ** (2 * (dim_t // 2) / self.num_pos_feats)
Traceback (most recent call last):
File "main.py", line 631, in
args = parser.parse_args()
File "main.py", line 533, in main
sampler_train.set_epoch(epoch)
File "/home/mmc_xiaolinhui/mmc_226_exp_202206/mdetr/engine.py", line 73, in train_one_epoch
loss_dict.update(criterion(outputs, targets, positive_map))
File "/home/mmc_xiaolinhui/anaconda3/envs/mdetr_env/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "/home/mmc_xiaolinhui/mmc_226_exp_202206/mdetr/models/mdetr.py", line 569, in forward
losses.update(self.get_loss(loss, outputs, targets, positive_map, indices, num_boxes))
File "/home/mmc_xiaolinhui/mmc_226_exp_202206/mdetr/models/mdetr.py", line 516, in get_loss
return loss_map[loss](outputs, targets, positive_map, indices, num_boxes, **kwargs)
File "/home/mmc_xiaolinhui/mmc_226_exp_202206/mdetr/models/mdetr.py", line 399, in loss_contrastive_align
beg_pos = tokenized.char_to_token(i, beg)
File "/home/mmc_xiaolinhui/anaconda3/envs/mdetr_env/lib/python3.8/site-packages/transformers/tokenization_utils_base.py", line 547, in char_to_token
raise ValueError("char_to_token() is not available when using Python based tokenizers")
ValueError: char_to_token() is not available when using Python based tokenizers
Start training...
Starting epoch 0
/home/mmc_xiaolinhui/mmc_226_exp_202206/mdetr/models/position_encoding.py:41: UserWarning: floordiv is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
dim_t = self.temperature ** (2 * (dim_t // 2) / self.num_pos_feats)
/home/mmc_xiaolinhui/mmc_226_exp_202206/mdetr/models/position_encoding.py:41: UserWarning: floordiv is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
dim_t = self.temperature ** (2 * (dim_t // 2) / self.num_pos_feats)
Traceback (most recent call last):
File "main.py", line 631, in
args = parser.parse_args()
File "main.py", line 533, in main
sampler_train.set_epoch(epoch)
File "/home/mmc_xiaolinhui/mmc_226_exp_202206/mdetr/engine.py", line 73, in train_one_epoch
loss_dict.update(criterion(outputs, targets, positive_map))
File "/home/mmc_xiaolinhui/anaconda3/envs/mdetr_env/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "/home/mmc_xiaolinhui/mmc_226_exp_202206/mdetr/models/mdetr.py", line 569, in forward
losses.update(self.get_loss(loss, outputs, targets, positive_map, indices, num_boxes))
File "/home/mmc_xiaolinhui/mmc_226_exp_202206/mdetr/models/mdetr.py", line 516, in get_loss
return loss_map[loss](outputs, targets, positive_map, indices, num_boxes, **kwargs)
File "/home/mmc_xiaolinhui/mmc_226_exp_202206/mdetr/models/mdetr.py", line 399, in loss_contrastive_align
beg_pos = tokenized.char_to_token(i, beg)
File "/home/mmc_xiaolinhui/anaconda3/envs/mdetr_env/lib/python3.8/site-packages/transformers/tokenization_utils_base.py", line 547, in char_to_token
raise ValueError("char_to_token() is not available when using Python based tokenizers")
ValueError: char_to_token() is not available when using Python based tokenizers