Tutorials

The object oriented structure of the repository can be daunting at first. To help new users get started, a tutorial notebook has been provided. This notebook walks through the basic functionality of the repository, including how to load data, create material objects, and access fits. The tutorial can be found in the ThermalModelTools folder of the repository as a Jupyter notebook named:

RepositoryTutorial.ipynb

This page will also describe some of the basic functions and methods for reference.

Let’s start by loading a material object from the save pickle file in the repository.

# Define the material of interest - make sure the name matches the folder name in the repository
material_of_interest = "Aluminum"
path_to_mat = os.path.join(<path_to_library>, material_of_interest)

mat_pickle_path = os.path.join(path_to_mat, f"material.pkl")
# Load the material object from the pickle file
with open(mat_pickle_path, "rb") as f:
    mat_obj = pkl.load(f)

Now that the material object is loaded, we can access its attributes and methods. Let’s practice by reviewing the available fits for this material.

[fit.name for fit in mat_obj.fits]

['Aluminum_data', 'Aluminum_1100_data', 'Aluminum_1100_NIST', 'Aluminum_3003F_NIST', ...]

We can also plot the data and fits using the built-in plotting methods.

# Plot the fits
mat_obj.plot_all_fits()

A common task is to calculate the thermal conductivity at a specific temperature using one of the fits. This can be done using the calc_tc method of the Fit class.

# Get thermal conductivity at a specific temperature
# First, select the fit to use
fit_to_use = mat_obj.fit_by_name("Aluminum_1100_data")
T = 50 * u.K
# Now use the calc_tc method
k, k_err = fit_to_use.calc_tc(T)

# Alternatively
# Load the fit directly from the material object
fit_to_use = mat_obj.fit_by_name("Aluminum_1100_data")
# Use the fit function and parameters directly
k = fit_to_use.function()(T.to(u.K).magnitude, *fit_to_use.parameters)
# This method might be used if you want to manipulate the output in some way
# and thus don't want to use the built in (more restrictive) calc_tc method.