medusa.analysis

Module Contents

class medusa.analysis.LinearRegression(add_intercept=True)

Linear regression model with normal equations, using PyTorch.

Parameters:

add_intercept (bool) – Whether to add an intercept term to the model

coef_

The model coefficients, including the intercept if applicable

Type:

torch.tensor

fit(X, Y)

Fits the model to the data (X, Y).

Parameters:

• X (torch.tensor) – The input data, with shape (n_samples, n_features)

• Y (torch.tensor) – The output data, with shape (n_samples, n_outputs)

predict(X)

Predicts the output for the given input data.

Parameters:

X (torch.tensor) – The input data, with shape (n_samples, n_features)

Returns:

Y_hat – The predicted output data, with shape (n_samples, n_outputs)

Return type:

torch.tensor

predict_from_range(range_=(- 1, 1), steps=100)

Predicts the output for the given input range.

Parameters:

• range (tuple) – The range of input values to predict over

• steps (int) – The number of steps to take between the range’s start and end

Returns:

Y_hat – The predicted output data, with shape (n_samples, n_outputs)

Return type:

torch.tensor