SoumiDas / train_driving

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Conditional Affordance Learning : https://arxiv.org/abs/1806.06498

  1. Please install the dependencies using environment.yml via anaconda3: conda env create -f environment.yml
  2. Once those are installed, activate the environment: conda activate CAL
  3. In order to run the training, run python training_cal.py
  4. I have kept only 3 episodes here as data: 2 for training, 1 for validation.
  5. This is going to run for 1 epoch only (the parameter is present in the last line of training_cal.py as the first argument of fit() )
  6. Intel Xeon 8 core, 32GB RAM, NVIDIA GTX 1080Ti 11GB takes around 33seconds for this particular setup.

About


Languages

Language:Python 100.0%