We had to work on a project with Swiss Plasma Center as part of the Machine Learning course CS-433 at EPFL.
- 95.43% Accuracy of classification.
-
Report: Contains the report in PDF. -
Python scripts:- 'data_creation.py': contains the code and libraries used for data generation and extraction.
- 'Diagnostic.py' and 'MANTIS.py': Auxiliairy material given by our project supervisor.
- 'model_training.py': contains the final CNN model we kept for Image Recognition. You can run the file on inputs of your choice in order to test also train the model.
-
Notebooks:- 'model_training.ipynb' : Since it is not possible to run the corresponding python script, we showed the results through this notebook.
- 'transfer_learning.ipynb' : Contains the methods applied for transfer learning and the results.