Tutorials

How to extract Light Curves

Assuming that the user has already applied the official Data Reduction pipeline over the data set that is to be studied and that the desired configurations have been already chosen, as specified in Section Configuration , one only has to run the following code, to get light curves from all of the stars.

from pyarchi import Photo_controller, store_data

controller = Photo_controller(job_number = 1,
                             config_path="<path_to_file>/config.yaml"
                             )
controller.optimize()  # not needed if the optimization process has already been done
data_fits = controller.run()

store_data(data_fits, job_number = 1, **controller.parameters)

Furthermore, if we want to change the configuration parameters of archi without having to edit the configuration file, we can do the following:

from pyarchi import Photo_controller, store_data

controller = Photo_controller(job_number = 1,
                             config_path="<path_to_file>/config.yaml"
                             )

configs_override = {'base_folder': "<new_path>",
                    "grid_bg": 0,
                    "initial_detect": 'dynam',
                    "method": "shape",
                    "detect_mode": "dynam",
                    "optim_processes": n_tasks,
                    "val_range": [1, 10],
                    "low_memory":0,
                    "fine_tune_circle":1,
                    "optimize":0,
                    'uncertainties': 1,
                    'CDPP_type': "K2",
                    "debug": 1,
                    "plot_realtime": 0,
                    'repeat_removal': 0}

# we can override all parameters from the file. Typically, one only needs to change these

controller.change_parameters(configs_override)
data_fits = controller.run()

store_data(data_fits, job_number = 1, **controller.parameters)

If we had a data set with bright and faint stars, we would have to change the “repeat_removal” parameter, to remove the brighter regions in the image. It is recommended to avoid more than 2 iterations, as using larger values may start recognizing regions of the background as stars.

In order to understand the organization of data inside the “data_fits” objects and the data that is stored with the “store_data” function, please refer back to Section Structure of the outputs.

It is highly recommended that the user builds its own routine to estimate the noise in the lightcurve. To do so, a given metric must be passed in the configs_override

This metric will be minimized during the mask selection routine. The current version only gives the user the flux and observation time.

from pyarchi import Photo_controller, store_data

controller = Photo_controller(job_number = 1,
                             config_path="<path_to_file>/config.yaml"
                             )
def noise_metric(flux, time):
    metric = np.std(flux)
    return metric
configs_override = {'CDPP_type': noise_metric}

How to plot the outputs

By default archi creates some plots, that are stored in the created folders, containing information of the light curves from each star and by comparing the target one against the DRP.

However, if we want to do so manually, we can do it this way:

from pyarchi import Photo_controller, store_data
import matplotlib.pyplot as plt

controller = Photo_controller(job_number = 1,
                             config_path="<path_to_file>/config.yaml"
                             )
controller.optimize()  # not needed if the optimization process has already been done
data_fits = controller.run()

for star in data_fits.stars: # iterate through all of the stars
    plt.plot(data_fits.mjd_time, star.photom)
    plt.title(f"{star.name}")
    plt.xlabel("MJD time [days]")
    plt.ylabel("Flux [ADU]")
    plt.show()

In order to better understand what information one can get from the controller, stars and data_fits, please refer back to relevant documentation.