vikrant7 / mobile-vod-bottleneck-lstm

Implementation of Mobile Video Object Detection with Temporally-Aware Feature Maps using PyTorch
131 stars 37 forks source link

train failed #3

Closed bryanlinnan closed 5 years ago

bryanlinnan commented 5 years ago

i tried to use your code to train, failed like below:

python train_mvod_basenet.py --datasets E:\ImageNetVID\ILSVRC2015 --batch_size 30 --num_epochs 10 --width_mult 1 2019-10-15 07:37:33,897 - root - INFO - Use Cuda. 2019-10-15 07:37:33,897 - root - INFO - Namespace(base_net_lr=None, batch_size=30, checkpoint_folder='models/', datasets='E:\ImageNetVID\ILSVRC2015', debug_steps=100, gamma=0.1, lr=0.003, milestones='80,100', momentum=0.9, num_epochs=10, num_workers=4, pretrained=None, resume=None, scheduler='multi-step', ssd_lr=None, t_max=120, use_cuda=True, validation_epochs=5, weight_decay=0.0005, width_mult=1.0) 2019-10-15 07:37:34,100 - root - INFO - Prepare training datasets. 2019-10-15 07:37:34,507 - root - INFO - using default Imagenet VID classes. 2019-10-15 07:37:34,507 - root - INFO - Stored labels into file models/vid-model-labels.txt. 2019-10-15 07:37:34,507 - root - INFO - Train dataset size: 76859 2019-10-15 07:37:34,507 - root - INFO - Prepare Validation datasets. 2019-10-15 07:37:34,569 - root - INFO - using default Imagenet VID classes. 2019-10-15 07:37:34,569 - root - INFO - <datasets.vid_dataset.ImagenetDataset object at 0x0000021307FEEE80> 2019-10-15 07:37:34,569 - root - INFO - validation dataset size: 11080 2019-10-15 07:37:34,569 - root - INFO - Build network. 2019-10-15 07:37:34,678 - root - INFO - Initializing weights of base net 2019-10-15 07:37:34,694 - root - INFO - Initializing weights of SSD 2019-10-15 07:37:39,750 - root - INFO - Learning rate: 0.003, Base net learning rate: 0.003, Extra Layers learning rate: 0.003. 2019-10-15 07:37:39,766 - root - INFO - Uses MultiStepLR scheduler. 2019-10-15 07:37:39,766 - root - INFO - Start training from epoch 0. Traceback (most recent call last): File "train_mvod_basenet.py", line 282, in device=DEVICE, debug_steps=args.debug_steps, epoch=epoch) File "train_mvod_basenet.py", line 115, in train for i, data in enumerate(loader): File "D:\Tool\Anaconda\Anaconda3-5.2.0\lib\site-packages\torch\utils\data\dataloader.py", line 819, in iter return _DataLoaderIter(self) File "D:\Tool\Anaconda\Anaconda3-5.2.0\lib\site-packages\torch\utils\data\dataloader.py", line 560, in init w.start() File "D:\Tool\Anaconda\Anaconda3-5.2.0\lib\multiprocessing\process.py", line 105, in start self._popen = self._Popen(self) File "D:\Tool\Anaconda\Anaconda3-5.2.0\lib\multiprocessing\context.py", line 223, in _Popen return _default_context.get_context().Process._Popen(process_obj) File "D:\Tool\Anaconda\Anaconda3-5.2.0\lib\multiprocessing\context.py", line 322, in _Popen return Popen(process_obj) File "D:\Tool\Anaconda\Anaconda3-5.2.0\lib\multiprocessing\popen_spawn_win32.py", line 65, in init reduction.dump(process_obj, to_child) File "D:\Tool\Anaconda\Anaconda3-5.2.0\lib\multiprocessing\reduction.py", line 60, in dump ForkingPickler(file, protocol).dump(obj) AttributeError: Can't pickle local object 'TrainAugmentation.init..'

D:\Work\Mobile Video Object Detection with Temporally-Aware Feature Maps\mobile-vod-bottleneck-lstm-master>2019-10-15 07:37:40,671 - root - INFO - Use Cuda. Traceback (most recent call last): File "", line 1, in File "D:\Tool\Anaconda\Anaconda3-5.2.0\lib\multiprocessing\spawn.py", line 105, in spawn_main exitcode = _main(fd) File "D:\Tool\Anaconda\Anaconda3-5.2.0\lib\multiprocessing\spawn.py", line 115, in _main self = reduction.pickle.load(from_parent) EOFError: Ran out of input

all dependency prepared Opencv Python3.6 Pytorch1.0

by the way, i work under windows

could you help tell what is the problem?

vikrant7 commented 5 years ago

Hi @bryanlinnan Since you were successful in training the model, I am closing this issue now.