Published May 2, 2026
| Version v1
Dataset
Open
Data supporting the results in the AutoREC paper
Authors/Creators
Description
Data Supporting the AutoREC Paper
This repository contains the data, trained models, and evaluation results used to support the results reported in the AutoREC paper.
Contents
- dataset/: Contains the EIS datasets used in the paper, including:
- training_dataset.pkl: Synthetic dataset used for model training.
- test_dataset.pkl: Synthetic dataset excluded from training and used for testing.
- battery_dataset.pkl: Experimental battery dataset.
- experimental_dataset.pkl: Additional experimental datasets, ordered as CO2 reduction, corrosion, and OER systems.
- models/: Contains the trained AutoREC agent used to generate the results reported in the paper. The folder includes:
- dqn_model.keras: Keras model file for the trained DQN agent.
- config.yaml: YAML file containing the training configuration and hyperparameters used for the DQN agent.
- training_reward.pkl: Pickle file containing the reward history during DQN training.
- no_deadloop/: Contains the corresponding files for a DQN agent trained without dead-loop mitigation.
- eval_results/: Contains the RL agent evaluation results for the EIS datasets, including:
- training.pkl: Evaluation results for the training dataset.
- test.pkl: Evaluation results for the test dataset.
- battery.pkl: Evaluation results for the battery dataset.
- experimental.pkl: Evaluation results for the additional experimental datasets.
- training_trials_0-9900/: Evaluation results for the training dataset across different training checkpoints.
- test_trials_0-9900/: Evaluation results for the test dataset across different training checkpoints.
Code repository
The AutoREC source code is available in the AutoREC GitHub repository.
The data in this repository are compatible with commit c63e63d.
Contact
For questions related to the data or calculation results, please contact:
- Yonatan Kurniawan: yonatan.kurniawan@utoronto.ca
Files
data_supporting_AutoREC_paper.zip
Additional details
Software
- Repository URL
- https://github.com/AUTODIAL/AutoREC
- Programming language
- Python
- Development Status
- Active