Cost functions module

Compute the cost of a trajectory.

pyrotor.cost_functions.compute_cost(trajectory, model, independent_variable)

Compute total cost of a trajectory given a quadratic model - Vectorized version.

Inputs:

• trajectory: DataFrame

  Trajectory of interest

• model: list of arrays or sklearn model

  Model of the cost; if list, the first element of the list is the constant c, the second one is the linear part w and the third one is the quadratic part q

• independent_variable: dict

  Describe the time-interval on which are defined the trajectories ex: {‘start’: 0, ‘end’: 1, ‘frequency’:.1}

Output:

• trajectory_cost: float

  Total cost of the trajectory

pyrotor.cost_functions.compute_cost_by_time(trajectory, model)

Compute the cost of a trajectory at every time given a quadratic model - Vectorized version.

Inputs:

• trajectory: DataFrame

  Trajectory of interest

• model: list of arrays or sklearn model

  Model of the cost; if list, the first element of the list is the constant c, the second one is the linear part w and the third one is the quadratic part q

Output:

• cost_by_time: ndarray

  Trajectory cost at each time

pyrotor.cost_functions.compute_f(vector_x, sigma_inverse, c_weight)

Compute the distance mean of a trajectory to the reference ones through its coefficients.

Inputs:

• vector_x: ndarray

  Coefficients of a single trajectory

• sigma_inverse: ndarray

  Pseudoinverse of the covariance matrix of the reference coefficients

• c_weight: ndarray

  Coefficients of a weighted trajectory

Output:

• cost: float

  Distance of the trajectory to the reference ones

pyrotor.cost_functions.compute_g(vector_x, format_model, extra_info)

Compute cost of a trajectory through its coefficients.

Inputs:

• vector_x: ndarray

  Coefficients of a single trajectory

• format_model: list of arrays or sklearn model

  Model of the cost; if list, the first element of the list is the the integrated linear part W and the second one the integrated quadratic part Q

• extra_info: dict

  Contains independent_variable, basis, basis_features and basis_dimension dictionaries

Output:

• g(vector_x): float

  The cost of the given trajectory

pyrotor.cost_functions.compute_trajectories_cost(trajectories, model, independent_variable)

Compute the cost for each trajectory of a list.

Inputs:

• trajectories: list of pd.DataFrame

  Each element of the list is a trajectory

• model: list of arrays or sklearn model

  Model of the cost; if list, the first element of the list is the constant c, the second one is the linear part w and the third one is the quadratic part q

• independent_variable: dict

  Describe the time-interval on which are defined the trajectories ex: {‘start’: 0, ‘end’: 1, ‘frequency’:.1}

Output:

• trajectories_cost: ndarray

  Array containing the cost of the trajectories

pyrotor.cost_functions.load_model(name)

Load model saved in a .pkl format.

Input:

• name: string

  Name of the .pkl file

Output:

• model: Python object

  Loaded machine learning model, scipy class…