This project mainly want to recongize the type of aurora shown in picture automatically by machine!
Reference: https://github.com/Flaick/Candev2018-aurora
Dataset: https://open.canada.ca/data/en/dataset/d700c863-8622-4ec2-a4ee-a1c377880e2e
Why class2 and class6: http://tid.uio.no/plasma/oath/
This project implements on the Pytorch framework.
‘class2’ and ‘class6’ folders contain the dataset used for training two different models.
‘model’ folder contains the final model for the class2 classification.
class2new.py file contains the procedure of building a model and training the model with dataset.
convert.py file contains the procedure of making our own dataset.
Depends on the time and memory of GPU, we only train one model for ‘class2’ which is saved in the ‘model’ folder
The model is built on the Resnet-18
Structure of training data:
└───Candev
├───class2
│ ├───candev_data_2_0
│ └───candev_data_2_1
└───class6
├───candev_data_6_0
├───candev_data_6_1
├───candev_data_6_2
├───candev_data_6_3
├───candev_data_6_4
└───candev_data_6_5