Optimizer

Fill in a module description here

---

source

Optimizer

 Optimizer (state:pandas.core.frame.DataFrame)

Optimize the number of points for t. Provided the number of points to sample in t based on t-1, return values number of sample points.

	Type	Details
state	DataFrame	a dataframe with the state of the administrative boundaries

class Optimizer:
    def __init__(self,
                 state:pd.DataFrame # a dataframe with the state of the administrative boundaries
                 ):
        "Optimize the number of points for t. Provided the number of points to sample in t based on t-1, return values number of sample points."
        self.state = state
        
        self.matrix = state.to_numpy()
        return

---

source

Optimizer.get_rank

 Optimizer.get_rank (is_benefit_x:list, w_vector:list, n_method:str=None,
                     c_method:str=None, w_method:str=None,
                     s_method:str=None)

Determines the rank of the administrative polygon based on the provided states.

@patch
def get_rank(self:Optimizer, 
             is_benefit_x:list,
             w_vector:list,  
            n_method:str=None,
            c_method:str =None, 
            w_method:str=None,
            s_method:str=None
            ):
    "Determines the rank of the administrative polygon based on the provided states."
    # normailize the matrix
    z_matrix, is_benefit_z = normalize(self.matrix, is_benefit_x, n_method)
    
     # replace nan values with 0 if cost and & 1 if benefit
    for i, is_benefit in enumerate(is_benefit_z):
        if is_benefit:
            z_matrix[:, i][np.isnan(z_matrix[:, i])] = 1
        else:
            z_matrix[:, i][np.isnan(z_matrix[:, i])] = 0

    if w_vector is None:
            # Weigh each criterion using the selected methods
            w_vector = weigh(z_matrix, w_method, c_method)
    else:
        pass

    s_vector, desc_order = score(z_matrix, is_benefit_z, w_vector, s_method)
    state = self.state.copy()
    state['value'] = s_vector
    df = state[['value']]

    # Get the indices of the sorted scores
    if desc_order:
        df_sorted = df.sort_values(by='value', ascending=False)
    else:
        df_sorted = df.sort_values(by='value', ascending=True)
    
    df_sorted['rank'] = range(1, len(df_sorted) + 1)
    df_rank = df_sorted[['rank']]
    
    return df_rank