Predict Photometric Redshift (Photo-Z)

This tool uses a random forest regression to predict photometric redshift of galaxies using features from PanSTARRS and unWISE survey data. It is tested against spectroscopic redshift estimates from SDSS survey. The training and testing of this model/data can be seen in the train_test directory, while the cross-matching of SDSS, PanSTARRS, and unWISE sources is shown in the xmatch directory.

The photoz_reg.pkl, main.py, example_data.csv, and predict_photoz.yaml files contain the saved model, the script to run the model, an example of data that can be run on this tool, and the enviornment on which this tool can run respectively.

How to use the tool:

1. Download photoz_reg.pkl, main.py,example_data.csv, and predict_photoz.yaml into any local directory.

2. In the terminal, navigate to the directory where the aformentioned files are stored.

3. Create an identical Conda environment by running conda env create -f predict_photoz.yaml in the terminal

4. Locate/download the following features of an object: ps_gpsfmag,ps_rpsfmag,ps_ipsfmag,ps_zpsfmag,ps_ypsfmag,ps_gkronmag,ps_rkronmag,ps_ikronmag,ps_zkronmag,ps_ykronmag,unwise_w1_mag_ab,unwise_w1_mag_vega,unwise_w2_mag_ab,unwise_w2_mag_vega

5. Create a .csv file that follows the template of example_data.csv with the aformentioned features OR simply revise example_data.csv to include new targets features. Make sure the .csv file with features is in the same directory as photoz_reg.pkl, main.py,example_data.csv, and predict_photoz.yaml.

6. Run main.py by running python main.py in the terminal

7. When prompted, enter the name of your file (e.g. example_data.csv) or press enter to run the tool with data from example_data.csv

8. Program will return the data of the features you entered and the predicted photometric redshift.

Results from running example_data.py should look like this:

✨ python main.py

Please enter the name of the data input folder: example_data.py

[Parallel(n_jobs=12)]: Using backend ThreadingBackend with 12 concurrent workers.

[Parallel(n_jobs=12)]: Done  26 tasks  | elapsed:  0.0s

[Parallel(n_jobs=12)]: Done 100 out of 100 | elapsed:  0.0s finished

ps_gpsfmag  19.296301

ps_rpsfmag  18.241699

ps_ipsfmag  17.846901

ps_zpsfmag  17.717300

ps_ypsfmag  17.301001

ps_gkronmag 18.182501

ps_rkronmag 17.292900

ps_ikronmag 16.792101

ps_zkronmag 16.573601

ps_ykronmag 16.399200

unwise_w1_mag_ab  16.471410

unwise_w1_mag_vega  13.772410

unwise_w2_mag_ab  16.933155

unwise_w2_mag_vega  13.594154

Name: 0, dtype: float64

Photo-Z Prediction: 0.09868804969999995