Photo-z analyses for the TPz algorithm

The objective of this project is to build a sample for TPz algorithm and analyze the results.

Manual for notebooks

The main folder is /IC_Photo-z/Iago. Here we have some files:

• Flexzboost.ipynb Notebook to execute FlexzBoost and analyse results.

• Controlled_sample.ipynb Notebook to build and analyze a sample with VIPERS and GAMA catalogs.

• Sanchez_run_TPz.ipynb Notebook to replicate the training sample from the paper (DOI: 10.1093/mnras/stu1836), analyse results and calculate weights.

• DR2-Spec_Querys.ipynb Notebook to build a sample using spectroscopic data from LIneA server.

• Plots.ipynb Notebook for plotting results; you can combine more than one result and compare.

Manual for samples folder

The path to the folder is: /IC_Photo-z/Iago/samples.

This folder contains all the training, validation and input files for ML algorithms.

Example: training.VVDS_VIPERS contains the training file built using VVDS and VIPERS catalog.

The folder /IC_Photo-z/Iago/results contains all results after running TPz.

REGRESSION: 0 CLASSIFICATION: 1 More than 1: tests

Example: results.GAMA_VIPERS_VVDS.1.mlz contains results from TPz using classification.

Manual for other files and folders

• IC_Photo-z/MLZ Folder with the TPz algorithm

• IC/Photo-z/Files png Contain all the images saved from notebooks

• /IC_Photo-z/Iago/samples/phot_sample_weighting_{catalog} Photometric data downloaded to weighting. Each data has the cuts used in spectroscopic associated.

• /IC_Photo-z/Iago/Sanchez_data Contains the data used to try to replicate Sanchez's paper results.