StructDepth icon indicating copy to clipboard operation
StructDepth copied to clipboard

some questions about training

Open mrwangxinqwe opened this issue 1 year ago • 0 comments

When I want to train, the default value given by val_split is nyu_test. I don't understand what the h5 file given in this file is.The following is the complete error message: loading model from folder model_weights/final_our Loading encoder weights... Loading depth weights... Loading pose_encoder weights... Loading pose weights... Training is using: cuda Models and tensorboard events files are saved to: logs/ Training is using frames: [0, -2, -1, 1, 2] Using train split: splits/nyu_train_0_10_20_30_40_21483-exceptfailed-21465.txt There are 21465 training items and -1 validation items

Using norm loss: True Using planar loss: True 0.05 for normloss 0.1 for planar loss vps_path: /home/wx/Gpan/output/ start epoch: 0 load weights folder: model_weights/final_our Training start from 0 epoch 0 step 0%| | 0/654 [00:00<?, ?it/s] Traceback (most recent call last): File "train.py", line 22, in trainer.train() File "/home/wx/Downloads/StructDepth-main/trainer.py", line 183, in train self.val() File "/home/wx/Downloads/StructDepth-main/trainer.py", line 637, in val for ind, (data, gt_depth, K, K_inv) in enumerate(tqdm(self.val_dataloader)): File "/home/wx/anaconda3/envs/StructDepth/lib/python3.6/site-packages/tqdm/std.py", line 1195, in iter for obj in iterable: File "/home/wx/anaconda3/envs/StructDepth/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 521, in next data = self._next_data() File "/home/wx/anaconda3/envs/StructDepth/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 1203, in _next_data return self._process_data(data) File "/home/wx/anaconda3/envs/StructDepth/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 1229, in _process_data data.reraise() File "/home/wx/anaconda3/envs/StructDepth/lib/python3.6/site-packages/torch/_utils.py", line 434, in reraise raise exception OSError: Caught OSError in DataLoader worker process 0. Original Traceback (most recent call last): File "/home/wx/anaconda3/envs/StructDepth/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 287, in _worker_loop data = fetcher.fetch(index) File "/home/wx/anaconda3/envs/StructDepth/lib/python3.6/site-packages/torch/utils/data/_utils/fetch.py", line 49, in fetch data = [self.dataset[idx] for idx in possibly_batched_index] File "/home/wx/anaconda3/envs/StructDepth/lib/python3.6/site-packages/torch/utils/data/_utils/fetch.py", line 49, in data = [self.dataset[idx] for idx in possibly_batched_index] File "/home/wx/Downloads/StructDepth-main/datasets/nyu_dataset.py", line 133, in getitem rgb, depth, norm, valid_mask = self.loader(line) File "/home/wx/Downloads/StructDepth-main/datasets/nyu_dataset.py", line 41, in h5_loader h5f = h5py.File(path, "r") File "/home/wx/anaconda3/envs/StructDepth/lib/python3.6/site-packages/h5py/_hl/files.py", line 408, in init swmr=swmr) File "/home/wx/anaconda3/envs/StructDepth/lib/python3.6/site-packages/h5py/_hl/files.py", line 173, in make_fid fid = h5f.open(name, flags, fapl=fapl) File "h5py/_objects.pyx", line 54, in h5py._objects.with_phil.wrapper File "h5py/_objects.pyx", line 55, in h5py._objects.with_phil.wrapper File "h5py/h5f.pyx", line 88, in h5py.h5f.open OSError: Unable to open file (unable to open file: name = 'data/nyu_test/00001.h5', errno = 2, error message = 'No such file or directory', flags = 0, o_flags = 0)

mrwangxinqwe avatar Sep 07 '24 04:09 mrwangxinqwe