Deep-learning for Ionogram Automatic Scaling

by Zhuowei Xiao, Jian Wang, Juan Li, Biqiang Zhao, Lianhuan Hu and Libo Liu

Code for paper 'Deep-learning for Ionogram Automatic Scaling'

Dataset

link to dataset: here.

link to trained model:

Google Drive here.

Baiduyun here Code: 2ftt

If you want to use your own data, you may first use 'GRM2HourlyRSF.m' and 'saveSAO2mat.m' matlab scripts to extract ionograms from GRM data. Then you may refer to jupyter notebooks under folder 'convert_GRM_to_input' to change data into desired format.

Prerequisites

• Python3.6
• Tensorflow2
• Segmentation-Models
• PyYaml

(If you are not familiar with python, we suggest you to use Anaconda to install these prerequisites.)

Installation

Clone this project to your machine.

git clone https://github.com/MrXiaoXiao/DIAS
cd DIAS

Training

You can use --gpu argument to specifiy gpu. To train a model, first create a configuration file (see example_config.yaml) Then run

python dias_main.py --train --gpu_id 0 --config-file YOUR_CONFIG_PATH

Tips: According to feedback that certain implementations of RAdam optimizer have problems in training convergence in this program, switch to Adam optimizer can solve the problem.

Testing

To test, run

python dias_main.py --test --gpu_id 0 --config-file YOUR_CONFIG_PATH

Evaluation

You can evaluate the model's performance by running script:

python dias_main.py --eval --config-file YOUR_CONFIG_PATH

Reference

If you find our work useful for your research, please consider citing: Zhuowei Xiao, Jian Wang*, Juan Li, Biqiang Zhao, Lianhuan Hu, Libo Liu. Deep learning for Ionogram Automatic Scaling. 2020. Advances in Space Research. https://doi.org/10.1016/j.asr.2020.05.009.

Contact

If you have any questions, please send email to jianwang@mail.iggcas.ac.cn or xiaozhuowei@mail.iggcas.ac.cn