cscribano / cerberus Goto Github PK

Traceback (most recent call last):
File "compressor.py", line 260, in
trainer_run()
File "compressor.py", line 211, in trainer_run
trainer.fit(model, train_loader, val_loader)
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 768, in fit
self._call_and_handle_interrupt(
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 721, in _call_and_handle_interrupt
return trainer_fn(*args, **kwargs)
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 809, in _fit_impl
results = self._run(model, ckpt_path=self.ckpt_path)
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1234, in _run
results = self._run_stage()
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1321, in _run_stage
return self._run_train()
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1343, in _run_train
self._run_sanity_check()
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1411, in _run_sanity_check
val_loop.run()
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/pytorch_lightning/loops/base.py", line 204, in run
self.advance(*args, **kwargs)
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/pytorch_lightning/loops/dataloader/evaluation_loop.py", line 154, in advance
dl_outputs = self.epoch_loop.run(self._data_fetcher, dl_max_batches, kwargs)
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/pytorch_lightning/loops/base.py", line 204, in run
self.advance(*args, **kwargs)
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/pytorch_lightning/loops/epoch/evaluation_epoch_loop.py", line 127, in advance
output = self._evaluation_step(**kwargs)
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/pytorch_lightning/loops/epoch/evaluation_epoch_loop.py", line 222, in _evaluation_step
output = self.trainer._call_strategy_hook("validation_step", *kwargs.values())
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1763, in _call_strategy_hook
output = fn(*args, **kwargs)
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/pytorch_lightning/strategies/strategy.py", line 344, in validation_step
return self.model.validation_step(*args, **kwargs)
File "compressor.py", line 118, in validation_step
preds = self.forward(img)
File "compressor.py", line 92, in forward
pred = self.backbone(img)
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "/home/inferno/Downloads/loconav/CERBERUS/models/cerberus.py", line 78, in forward
scn_out = self.head_scn(small, argmax=inference)
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "/home/inferno/Downloads/loconav/CERBERUS/models/heads.py", line 96, in forward
w_pred, s_pred, t_pred = torch.split(x, self.cls_splits, 1)
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/torch/functional.py", line 189, in split
return tensor.split(split_size_or_sections, dim)
File "/home/inferno/anaconda3/envs/cerberus/lib/python3.8/site-packages/torch/_tensor.py", line 611, in split
return super(Tensor, self).split_with_sizes(split_size, dim)
RuntimeError: start (14) + length (4) exceeds dimension size (16).

Thank you for the author's contribution. In this framework, It was very easy and effective to use the BDD100K open dataset for training. Therefore, I would like to further fine-tune the model using actual experimental scene data. The annotation work for object detection and scene classification can be done easily in the format of BDD100K.
However, I would like to ask for the author's suggestion on how to annotate road lines for lane detection in a proper way?

Hi, looking for a quantized model if you can make avaialble

Traceback (most recent call last):
File "inference/run.py", line 19, in
from dataset.utils.transforms import Preproc
ModuleNotFoundError: No module named 'dataset'

Hi cant find the pth files in the link,kindly fix