In this project, we are going to solve a classification problem with 5 classes of (O-C-S rich) AGBs, Post AGBs and Young Stellar Objects (YSOs) by their spectroscopy data. The raw data which is different classes' data files are provided in the main separately.
Milestone | Report | Notebook(s) |
---|---|---|
Data | Report 1 | |
Traditional ML Techniques | Report 2 | |
Neural Networks | Report 3 |
-
In the first Milestone we have cleaned up the data and normalized the features by Standard, Robust and MinMax scalers. We have also encoded the output classes by One-Hot Encoding method and you can find all of the aforementioned data files in 1st Milestone Data.
-
The first Milestone's report and its (ipynb) code file are provided in first Milestone.
-
First milestone files all together: Milestone_1.zip.
- Applying the traditional techniques (SGD, KNN, SVM, etc.) to the data, Tunning Models and their Evaluation.
- Applying a Neural Network model to the data.