Alma-bachelor-project

This project was completed as a part of our Bachelor's thesis. The project was carried out during the second half of the academic year of 2022-2023. This project provides a classifier for ALMA (Atacama Large Millimeter/submillimeter Array) data, specifically for the purpose of detecting protoplanetary disks in the data.

Authors

• @A1

Acknowledgements

• AK1
• AK2
• AK3

Badges

Installation

Install the project with:

  # TODO

Documentation

Documentation

FAQ

Question 1

Answer 1

Question 2

Answer 2

Features

• F1
• F2
• F3

License

MIT

Project structure

├── alma-classifier/
│   ├── __init__.py
│   ├── alma_classifier/
│   │   ├── __init__.py
│   │   │
│   │   ├── data_acquisition/
│   │   │   ├── __init__.py
│   │   │   ├── alminer_mod.py
│   │   │   ├── constants_copy.py
│   │   │   └── mine_n_fits.py
│   │   ├── image_processing/
│   │   │   ├── __init__.py
│   │   │   ├── image_augmentation.py
│   │   │   └── pre_processing.py
│   │   ├── Imager.py
│   │   ├── models/
│   │   │   ├── __init__.py
│   │   │   ├── pytorch/
│   │   │   │   ├── __init__.py
│   │   │   │   └── model_01/
│   │   │   │       ├── __init__.py
│   │   │   │       ├── constants.py
│   │   │   │       ├── evaluation.py
│   │   │   │       ├── model.py
│   │   │   │       └── training.py
│   │   │   └── tensorflow/
│   │   │       ├── __init__.py
│   │   │       └── model_01/
│   │   │           ├── __init__.py
│   │   │           ├── constants.py
│   │   │           ├── evaluation.py
│   │   │           ├── model.py
│   │   │           └── training.py
│   │   ├── pipeline_pytorch.py
│   │   ├── pipeline_tensorflow.py
│   │   └── utils/
│   │       └── __init__.py
│   ├── config.py
│   ├── data/
│   │   ├── datasets/
│   │   │   ├── raw/
│   │   │   │   ├── neg/
│   │   │   │   └── pos/
│   │   │   └── reviewed/
│   │   │       ├── test/
│   │   │       │   ├── neg/
│   │   │       │   └── pos/
│   │   │       ├── train/
│   │   │       │   ├── neg/
│   │   │       │   └── pos/
│   │   │       └── val/
│   │   │           ├── neg/
│   │   │           └── pos/
│   │   └── fits/
│   │       ├── neg/
│   │       └── pos/
│   ├── docs/
│   │   └── branches.md
│   ├── LICENSE
│   ├── notebooks/
│   │   ├── data_acquisition.ipynb
│   │   ├── model_evaluation.ipynb
│   │   └── model_training.ipynb
│   ├── requirements.txt
│   └── tests/
│       ├── test_data_acquisition.py
│       ├── test_model_pytorch.py
│       ├── test_model_tensorflow.py
│       └── test_pre_processing.py

Structure Explained

alma-classifier/

This directory contains the source code for the ALMA classifier.

alma_classifier/

This directory contains the implementation of the ALMA classifier.

data_acquisition/

This directory contains the implementation for data acquisition.

alminer_mod.py: implementation of Alminer class that mines the data from the FITS files. constants_copy.py: constant variables for data acquisition. mine_n_fits.py: implementation of mine_n_fits function that reads the FITS files.

image_processing/

This directory contains the implementation for image processing.

image_augmentation.py: implementation of image augmentation. pre_processing.py: implementation of image pre-processing.

models/

This directory contains the implementation of the neural network models.

pytorch/: implementation of the PyTorch model. tensorflow/: implementation of the TensorFlow model.

utils/

This directory contains utility functions.

Other Files

Imager.py: implementation of the Imager class. pipeline_pytorch.py: implementation of the PyTorch pipeline to keep model sandboxed. pipeline_tensorflow.py: implementation of the TensorFlow pipeline. config.py: configuration file with, if needed, global settings for whole project LICENSE: license file.

data/

This directory is where all data used in the project is stored. It is divided into two subdirectories:

datasets/

This directory contains the datasets.

raw/: raw genererated training data. reviewed/: reviewed raw data.

fits/ This directory contains the FITS files.

Other Directories/files docs/: documentation. notebooks/: This directory contains Jupyter notebooks used for exploratory data analysis and experimentation. tests/: unit tests. reports/: This directory contains any reports generated by the project, including visualizations, analysis, and findings. requirements.txt: This file lists all the Python packages required to run the project. setup.py: This file contains the project metadata required for packaging the project as a Python distribution.

Workflow

Branches # TODO