71b4611c64
Version v1 generates a better baseline with higher max_episodes and reward_threshold attained.
changed_params -->
register(
id='CartPole-v1',
entry_point='gym.envs.classic_control:CartPoleEnv',
max_episode_steps=500,
reward_threshold=475.0,
)
|
||
|---|---|---|
| .. | ||
| basic_examples | ||
| domain_templates | ||
| README.md | ||
| __init__.py | ||
| bug_report_model.py | ||
| run_ddp-examples.sh | ||
| run_examples-args.sh | ||
| test_examples.py | ||
README.md
Examples
Our most robust examples showing all sorts of implementations can be found in our sister library lightning-bolts.
Basic examples
In this folder we add 3 simple examples:
- MNIST Classifier (defines the model inside the
LightningModule). - Image Classifier (trains arbitrary datasets with arbitrary backbones).
- Autoencoder (shows how the
LightningModulecan be used as a system)
Domain examples
This folder contains older examples. You should instead use the examples in lightning-bolts for advanced use cases.