Skip to content

TensorFlow models repository with RALI support

Notifications You must be signed in to change notification settings

sampath1117/RALImodels

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 

Repository files navigation

RALImodels

TensorFlow models repository with RALI support

Steps to run SSD detection training for COCO2017 on a Rocm3.5 + TensorFlow 1.15 docker container:

Outside the container:

cd $HOME
mkdir dockerMount
git clone https://github.com/r-abishek/RALImodels.git
sudo docker pull abishekr/mlperf_rocm3.5_tf1.15:v0.1.5
sudo docker run -it --network=host --memory=14G --device=/dev/kfd --device=/dev/dri --ipc=host --shm-size 16G --group-add video --cap-add=SYS_PTRACE --security-opt seccomp=unconfined -v $HOME/dockerMount:/media abishekr/mlperf_rocm3.5_tf1.15:v0.1.5

Inside the container:

cd /media/RALImodels/models/research
./download_all.sh /media/ssdTraining/data/coco2017_tfrecords /media/ssdTraining/checkpoints
bash ./examples/SSD320_FP16_1GPU.sh /media/ssdTraining/checkpoints/ &> /media/ssdTraining/checkpoints/log.txt

The log is generated in log.txt.

About

TensorFlow models repository with RALI support

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 93.2%
  • Jupyter Notebook 5.1%
  • Shell 1.3%
  • Other 0.4%