Unverified Commit a40e67f1 authored by MasterScrat's avatar MasterScrat Committed by GitHub

Merge pull request #2 from metataro/wandb

Updated README
parents 6813dbd1 0f4b590d
# NeurIPS 2020 Flatland Challenge baselines
The basic structure of this repository is adopted from [https://github.com/spMohanty/rl-experiments/](https://github.com/spMohanty/rl-experiments/)
📈 [**Results**](https://app.wandb.ai/masterscrat/flatland/reports/Flatland-Baselines--Vmlldzo4OTc5NA)
## Installation
Experiments
---
Experiments consist of one or many RLlib YAML config files alongside a MARKDOWN file containing results, plots and a detailed description of the methodology.
All files are stored in a experiment folder under `experiments/<env-name>/<experiment-name>`.
- [Tree observations w/ fully connected network](experiments/flatland_random_sparse_small/tree_obs_fc_net)
- [Global observations w/ convnet](experiments/flatland_random_sparse_small/global_obs_conv_net)
Setup
---
Tested with Python 3.6 and 3.7
......@@ -10,7 +21,7 @@ Tested with Python 3.6 and 3.7
conda create --name ray-env python=3.7 --yes
```
You may need to install/update bazel: https://docs.bazel.build/versions/master/install-ubuntu.html
You may need to install/update bazel: [Ubuntu guide](https://docs.bazel.build/versions/master/install-ubuntu.html)
```
pip install ray[rllib]
......@@ -19,23 +30,18 @@ pip install -r requirements.txt
```
## Usage
```
Training example:
python ./train.py -f experiments/flatland_random_sparse_small/global_obs_conv_net/ppo.yaml
Test example:
python ./rollout.py /tmp/ray/checkpoint_dir/checkpoint-0 --run PPO --no-render
--config '{"env_config": {"test": true}}' --episodes 1000 --out rollouts.pkl
`python ./train.py -f experiments/flatland_random_sparse_small/global_obs_conv_net/ppo.yaml`
Note that -f overrides all other trial-specific command-line options.
```
Test example:
## Experiment structure
`python ./rollout.py /tmp/ray/checkpoint_dir/checkpoint-0 --run PPO --no-render
--config '{"env_config": {"test": true}}' --episodes 1000 --out rollouts.pkl`
Experiments consist of one or many rllib YAML config files
alongside a MARKDOWN file containing results, plots
and a detailed description of the methodology.
All files are stored in a experiment folder under `experiments/<env-name>/<experiment-name>`.
An example can be found under `experiments/flatland_random_sparse_small/global_obs_conv_net/README.md`.
Note that -f overrides all other trial-specific command-line options.
Notes
---
- The basic structure of this repository is adopted from [https://github.com/spMohanty/rl-experiments/](https://github.com/spMohanty/rl-experiments/)
\ No newline at end of file
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment