How to train driving dataset on timesformer?

facebookresearch / TimeSformer

The official pytorch implementation of our paper "Is Space-Time Attention All You Need for Video Understanding?"

Other

1.51k stars 209 forks source link

How to train driving dataset on timesformer? #119

Open pgupta119 opened 1 year ago

pgupta119 commented 1 year ago

Hi Team, I am using the driving dataset to understanding road behavior. I would like to train the transformer with the predefine class such as ['left turn', 'right turn', 'U turn']. Then I would like to test my data on the Timesformer. I would like to know how to train the timesformer.

syamamo1 commented 1 year ago

Read the README.md. And DATASET.md. You'll end up running something like python tools/run_net.py \ --cfg configs/Kinetics/TimeSformer_divST_8x32_224.yaml \

Make a .yaml config file and be sure to specify finetuning and the number of classes/batch size/num GPU