Using JWST-NIRISS F115 & F120W grism spectroscop to identify quiescent galaxies, selected on morphology via supervised machine-learning trained upon SDSS imaging and validate with independent JWST
Prerequisites
- Python 3
- Git
- Clone the repository
git clone https://github.com/it-guy007/JWST-passage-Quiescent-galaxies.git
cd JWST-passage-Quiescent-galaxies- Create a virtual environment
python -m venv venv
source venv/bin/activate # On Windows use `venv\Scripts\activate`- Install the requirements
pip install -r requirements.txt- Create the folder structure:
JWST-passage-Quiescent-galaxies
├── data
│ ├── images_training_rev1
│ ├── training_solutions_rev1.csv
│ └── converted
│
├── tf_data
└── models
- Download the data from the Kaggle competition Galaxy Zoo - The Galaxy Challenge and place it in the data folder
Alternatively, you can use the Kaggle API to download the data:
kaggle competitions download -c galaxy-zoo-the-galaxy-challenge -p data/
unzip data/galaxy-zoo-the-galaxy-challenge.zip -d data/The notebook has all the logic required for creating the model. Modify the needed variables for needed extra configuration.
The notebook has all the logic for evaluating the model and creating the needed plots for analysis.
The notebook has all the logic for using the model to create predictions of fits files. It generates a csv with the predictions stored for further analysis and if activated a separate directory with the images desired above a certain accuracy.
The notebook has all the logic for evaluating the predictions and creating the needed plots for analysis.
It is also possible to download a pre-trained model for this project, which are the steps done in the builder notebook. The model is available at Hugging Face. Move both files in data to the local model folder of the project.
This project is licensed under the MIT License see the LICENSE file for details.