Issues
- 2
visualization
#52 opened by Traveler-Pass - 0
visualization
#73 opened by fengjingchehu - 3
train new dataset with pretrained model
#69 opened by sanwei111 - 0
- 6
decord._ffi.base.DECORDError: [16:56:24] /github/workspace/src/video/video_reader.cc:151: Check failed: st_nb >= 0 (-1381258232 vs. 0) ERROR cannot find video stream with wanted index: -1
#24 opened by NEUdeep - 0
About using pre-trained models
#71 opened by Manav2906 - 0
about the beginning index of the annotations
#70 opened by sanwei111 - 0
frames*crops*clips and train_log
#68 opened by sanwei111 - 1
Frames x Crops x Clips
#67 opened by sanwei111 - 1
about how to train
#65 opened by sanwei111 - 1
about data input
#66 opened by sanwei111 - 0
Sudden slowdown in training
#63 opened by tss35 - 0
About hmdb51 Dataset
#64 opened by yyhyyh17 - 2
训练数据集为什么不打乱?
#62 opened by Turing77 - 1
something something V2's fps
#61 opened by pudding126 - 1
- 1
- 1
- 7
BatchSize How to set in a single card?
#39 opened by anfengmin100 - 1
How to rerun this project in ucf101?
#56 opened by sja9527 - 4
Problem of testing
#55 opened by xiaofengWang-CCNU - 10
Problems related to model performance
#51 opened by jwfanDL - 1
can't find file
#54 opened by Brasiletto - 1
Question of setting of 'new_length' in dataset.py
#48 opened by dy1998 - 1
Backward and forward in L-TDM module
#50 opened by aikuniverse - 1
- 2
Can't reproduce kinetic400 result
#49 opened by sean186 - 5
Fine tuning of UCF101 and HMDB51
#28 opened by LV0913 - 2
Some confusion
#46 opened by wanlmq6165163 - 1
- 2
About the reproduction of the hmdb51 dataset
#45 opened by Brasiletto - 3
- 2
dataset.py 'new_length'
#42 opened by zhanggj821 - 1
About memory of training on sthv1
#43 opened by Oldyin1012 - 1
TypeError: 'Compose' object is not iterable
#41 opened by venje-zhu - 3
Question about TSM accuracy in ablation study
#40 opened by jinhaoduan - 0
Batch_Size H欧文、 to set size in a single card
#38 opened by anfengmin100 - 1
Confusion about the baseline
#37 opened by wei666666 - 1
sth官网没了,能否分享一下sth-sth-v1的原始csv格式的label文件呢?
#36 opened by takfate - 1
Visualization wtih Grad-CAM
#35 opened by wei666666 - 0
- 1
A question about testing
#33 opened by Brasiletto - 2
When testing on sth-sth-v1, which checkpoint should be used, 'best.pth' or '60_epoch_ckpt_pth'?
#31 opened by withbrightmoon - 0
CUDA out of memory, even when batch_size=1
#32 opened by fightingaaa - 1
Question about processing of HMDB51 dataset
#29 opened by MonkeyKing-KK - 6
Pretrained models
#27 opened by LV0913 - 5
- 1
- 1
- 1