Open Zhuysheng opened 6 years ago
Hi @Zhuysheng , are you able to provide more details, or a minimal code that reproduces the error? Thanks!
@chaoyuaw first, i use reencode.sh to create the mpeg4_videos ,
then run
python train.py --lr 0.0003 --batch-size 40 --arch resnet152 \
--data-name ucf101 --representation iframe \
--data-root data/ucf101/mpeg4_videos \
--train-list data/datalists/ucf101_split1_train.txt \
--test-list data/datalists/ucf101_split1_test.txt \
--model-prefix ucf101_iframe_model \
--lr-steps 55 110 165 --epochs 220 \
--gpus 0 1
after print the net, it shows could not open input stream,
Training arguments: data_name: ucf101 data_root: data/ucf101/mpeg4_videos train_list: data/datalists/ucf101_split1_train.txt test_list: /home/zys/pytorch-coviar/data/datalists/ucf101_split1_test.txt representation: iframe arch: resnet152 num_segments: 3 no_accumulation: False epochs: 220 batch_size: 40 lr: 0.0003 lr_steps: [55.0, 110.0, 165.0] lr_decay: 0.1 weight_decay: 0.0001 eval_freq: 5 workers: 8 model_prefix: ucf101_iframe_model gpus: [0, 1]
Initializing model: base model: resnet152. input_representation: iframe. num_class: 101. num_segments: 3.
Model( (base_model): ResNet( (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False) ........ ) Augmentation scales: [1, 0.875, 0.75, 0.66] Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream ^Z [2]+ Stopped
@Zhuysheng hey,i met same error,have you figure it out yet ?
@chaoyuaw @FelixZhang7 sorry, i havent solve it ,any idea?
maybe you should check the video used for train or test. you can print the video name in data_loader/coviar_data_loader.c to get which video can not be opened
@Zhuysheng hey,i met same error,have you figure it out yet ?
i follow the GETTING_STARTED.md, but when run train.py : Augmentation scales: [1, 0.875, 0.75, 0.66] Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream Could not open input stream