error for video format when training kinetics from scratch

[InwoongLee]

I'm trying to train kinetics from scratch. I finished caffe2 from source with USE_OPENCV=1 USE_FFMPEG=1 USE_LMDB=1. However, I received these errors.

[E video_decoder.cc:150] Insufficient data to determine video format [E video_decoder.cc:751] The video seems faulty and we could not decode enough frames: 0 VS 8 INFO:train_net:Finished iteration 0/1000000 of epoch 0 (0.65 clips/sec) INFO:train_net:Training loss: 5.99146461487, lr: 0.00999999977648, accuracy: 0.0 [E videodecoder.h:269] Error opening video file ^@^@^@^\ftypisom^@^@^B^@isomiso2mp41^@^@^@^Hfree^@?3�mdat�^D^@Lavc57.107.100^@BV^���8^@~H��^X:^Z^^~I~I�`�~X:G~K��ܷ^Nt��޺~Q�2�~K�/n���gpw���~W^^��^Y"^N~IU�^S��T�h~J^H�~HG~N�^E1^Dt;^N@����M�y~��~G^Vs^D�� �^�f~Rc^��=6��~L'�z�/^Y��^G�� �La^H��Wn���2~I�B~_G�^\�^X,g?>ܶ ^Q^N^Ln�s�9G6x~L^Z^Rl�^Q6l#^@~^pqf^?

I don't know how to error for opening video file. Anybody knows this?

[dutran]

This is either your videos has no frame or the parameter setting is incorrect. Please pay attention to use_list (in create_video_db.py) and use_local_file (in train_net.py). These parameters are needed to be used together, meaning both be 0 or 1.

[InwoongLee]

I set use_list to 0 in create_video_db.py and use_local_file in train_net.py. Actually, that is the default settings. Can you guess this problem's reason? Thank you for your reply.

[dutran]

From you error message, it looks like you use use_local_file=1 that's why it said a weird filename in the log. To debug, you may want to try again with both use_list=1 and use_local_file=1