Moose

This is an official pytorch implementation of MOOSE: Pay Attention to Temporal Dynamics for Video Understanding via Optical Flows

pip 
python setup.py build develop & pip install e .
CUDA_VISIBLE_DEVICES=2 python tools/run_net.py --cfg /data2/hongn/TimePSFormer/configs/Kinetics/Moose_divST_8x32_224.yaml

Code base was inspired by TimeSFormer Colab examples and runs: TBA

Name		Name	Last commit message	Last commit date
configs		configs
slurm_scripts		slurm_scripts
timesformer		timesformer
tools		tools
.gitignore		.gitignore
CODE_OF_CONDUCT.md		CODE_OF_CONDUCT.md
CONTRIBUTING.md		CONTRIBUTING.md
LICENSE		LICENSE
README.md		README.md
SSv2-haa500-AVA-preprocess.ipynb		SSv2-haa500-AVA-preprocess.ipynb
Sapiens-load-test.ipynb		Sapiens-load-test.ipynb
TimeSFormer-timeonly.ipynb		TimeSFormer-timeonly.ipynb
environment.yml		environment.yml
example.ipynb		example.ipynb
extract-ssv2-frames.sh		extract-ssv2-frames.sh
filter-none-videos-dataset.ipynb		filter-none-videos-dataset.ipynb
load-moose.ipynb		load-moose.ipynb
requirement.txt		requirement.txt
setup.cfg		setup.cfg
setup.py		setup.py
test.csv		test.csv
test_mamba.ipynb		test_mamba.ipynb
test_mamba.py		test_mamba.py
train.csv		train.csv
val.csv		val.csv
vision_transformer.py		vision_transformer.py
viz_time_space.ipynb		viz_time_space.ipynb

Navigation Menu

Repository files navigation

Moose

About

Releases

Packages

Used by

Contributors

Languages

Latest commit

History

Folders and files

Repository files navigation

Moose

About

Resources

Code of conduct

Contributing

Stars

Watchers

Forks

Releases

Packages

Used by

Contributors

Languages