Here alg = dreamer,dreambc
Training on walker-walk: python alg_main.py --env walker-walk
Task Generalization on walker-stand: python alg_main.py --env walker-stand --model results/walker-walk/alg/models.pth
Feature Generalization on walker-walk: python alg_main.py --env walker-walk --img-source color --id alg-color
Render Episodes: python alg_main.py --env walker-walk --model results/walker-walk/alg/models.pth --agent-model results/walker-walk/alg/models.pth --test --test-episodes 5 --video
TODO: Switch to matplotlib .png from plotly .html Add functionality for multiple runs Rerun experiments with multiple seeds