highway-environment-nn-model-training
https://github.com/gordanliang/highway-environment-nn-model-training
Science Score: 44.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (6.8%) to scientific vocabulary
Repository
Basic Info
Statistics
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
System Requirement
windows : tensorflow=2.13.0
ubuntu 18.04 : tensorflow=2.13.1
Install
First git clone
bash
git clone https://github.com/gordanLiang/highway-environment-NN-model-training.git
cd to file
bash
cd highway-environment-NN-model-training
Using conda environment python==3.8
bash
conda create -n <env name> python=3.8
activate environment
bash
conda activate <env name>
intstall requirements with
bash
pip install -r requirements.txt
Training DQN model with stable baseline 3
Training dqn model using
bash
python highway_dqn_train.py
Testing dqn model using
bash
python highway_dqn_test.py
It will give 10 score each test 10 rounds and collect reward.
Generating dataset with pre-trained dqn model
Load model and generate dataset in highway environment by using
bash
python dataset_making_each1w.py
It will generate total 50000 data for each action 10000 data
Training NN model using dataset
Before generating dataset, we can use it to train the NN model and test it in highway environment.
Training the NN model using
bash
python NN_model_train.py
Testing in highway environment
bash
python NN_model_test.py
This is the same way tesing dqn model so can compare.
Result
DQN 10 reward
bash
[212, 160, 229, 227, 207, 235, 227, 219, 171, 230]
average:211.7
NN model 10 reward
bash
[229, 227, 222, 214, 189, 241, 212, 194, 222, 198]
average:214.8
References
highway envirenment:https://github.com/Farama-Foundation/HighwayEnv
stable baseline 3:https://github.com/DLR-RM/stable-baselines3
Owner
- Login: gordanLiang
- Kind: user
- Repositories: 1
- Profile: https://github.com/gordanLiang
Citation (CITATION.cff)
cff-version: 1.2.0 message: "If you use this software, please cite it as below." authors: - family-names: "Leurent" given-names: "Edouard" title: "An Environment for Autonomous Driving Decision-Making" version: 1.4 date-released: 2018-05-01 url: "https://github.com/eleurent/highway-env"