Getting Started with AstroGeoFit

Python Version License

Warning

In order to have an easy first contact with AstroGeoFit, we provide a small example which containes synthetic data, which we used to test the code while developing it. Here you can find a step-by-step guide on how to execute this example and visualize the results.

Observe the Files Found on the Folder

The first important thing to do is to do a quick check on the files that the software uses. To do this please accede to the examples/synthetic_data folder. There you will find three folders : configuration_file, data and results.

The folder configuration_file contains all the needed variables that the tool uses as base configuration. For this example, nothing has to be modified, but if it can be used as “sandbox” to get a first idea of how the program uses the file. A full guide on what are these variables and how to set them can be found in Configuration File Guide.

The folder data containes the data set that will be used for this example as well as the age-depth model and the eccentricity solution used.

The folder results is empty and will be used to store the results of the execution.

Execute the Fitting and MCMC of the Example

Note

  • The folder examples has to bee in the same directory as the AstroGeoFit_tool.py file. If not, this will not work.

To execute the basic example, first open a terminal (bash) in the main folder of the package, where you can find the AstroGeoFit_tool.py file. Once it is open, execute the following command:

python AstroGeoFit_tool.py --basic_example --fit --mcmc

This will execute the genetic algorithm fitting and the MCMC calcula using the synthetic data set. If you see the message:

INFO -  All parameters obatined, starting analysis.

Means that the execution started correctly. Now a progress bar will appear.

Once the message

INFO - Data updated and saved to ./examples/synthetic_data/results/syntehtic_data_results_mcmc.pickle

appears on the terminal, the execution of the fitting and the MCMC will be terminated.

Visualize the Results Obtained

To visualize the results obtained the notebooks provided can be used. The AstroGeoFit_Analysis_Model_Fitting notebook will display the results obtained by the genetic algorithm, while the AstroGeoFit_Analysis_MCMC will display the results of the MCMC. To open these notebooks open a terminal in the main folder of AstroGeoFit and execute the folliwng command:

jupyter notebook

When executing this, a navigator (chrome, mozilla, safari, etc.) should open. If not, in the terminal will appear a link similar to this:

http://localhost:8888/tree?token=bf9f95f8bb2d05ce5c7988287bd91a23e0e313d3ee2db8e2

Copy this link into a browser window and press enter.

Now you should see a folder with all the files that exist in the AstroGeoFit folder. Open the necessary notebook.

In order to obtain the correct results it is necessary to change the variable configuration_file_path found in the notebook in the following way:

configuration_file_path = "basic_example"

Once this is done, the notebook can be executed. In order to find a guide which explains the figures in detail, please refer to Visualization of the Results Guide

Warning

  • In the fitting notebook there is a plot that shows the significance test results. As we did not execute it this time, this will not work. It is best to not execute this cell.

Next Steps

Now a first execution of the AstroGeoFit tool has been done it is time to explore in more detail the rest of the tool. The next step could be either try to execute one of the examples found on the paper which can be found in Obtention of the Paper Results, learn how to adapt the configuration file to your data by following the guide found in Configuration File Guide or learn the different execution options that AstroGeoFit proportionates by following the guide found in Execution of the Package.