README.md

File Introduction

bigtransfer.py original one bigtransfer1.py original one bigtransfer2.py change zero grad bigtransfer3.py change optimizer bigtransfer4.py and bigtransfer5.py distributed training

work1.sbatch run original one work2.sbatch change zero grad work3.sbatch change optimizer work4.sbatch run with 2 gpu work5.sbatch run with 4 gpu

Run Experiment

sbatch work1.sbatch sbatch work2.sbatch sbatch work3.sbatch sbatch work4.sbatch sbatch work5.sbatch