oracle.pretrained.BTS

Pretrained model(s) for the BTS dataset.

Functions

augment_table(table)

Augments a table by modifying its time-related and feature-specific values, and splitting it into two separate tables. This function performs the following modifications: - Converts filter IDs ('fid') to their corresponding mean wavelengths using the mapping ZTF_fid_to_wavelengths. - Normalizes the 'jd' column by subtracting the minimum jd value to set the starting time at zero. - Reorders the columns based on a predefined list time_dependent_feature_list to create a time-dependent table, and adds a constant column 'flag' with a value of 1. - Extracts a time-independent table based on a predefined list time_independent_feature_list. :param table: The input table containing astronomical observations with at least 'fid' and 'jd' columns. :type table: pandas.DataFrame.

Classes

ORACLE1_BTS([model_dir])

ORACLE1_BTS is a model class that inherits from GRU_MD designed to load a pre-trained BTS model and perform predictions on time series data augmented with static features.