metrics.py

import numpy as np
import pandas as pd

from sklearn.metrics import confusion_matrix


def rmse(y_hat, y):
    return np.sqrt(np.mean(np.square(y_hat-y)))


def accuracy(y_hat, y):
    """
    Function to calculate the accuracy

    Inputs:
    > y_hat: pd.Series of predictions
    > y: pd.Series of ground truth
    Output:
    > Returns the accuracy as float
    """
    """
    The following assert checks if sizes of y_hat and y are equal.
    Students are required to add appropriate assert checks at places to
    ensure that the function does not fail in corner cases.
    """
    assert(y_hat.size == y.size)
    assert(y_hat.size > 0)
    # TODO: Write here
    return (y_hat == y).sum()/y.size


def precision(y_hat, y, cls):
    """
    Function to calculate the precision

    Inputs:
    > y_hat: pd.Series of predictions
    > y: pd.Series of ground truth
    > cls: The class chosen
    Output:
    > Returns the precision as float
    """
    assert(y_hat.size == y.size)
    assert(y_hat.size > 0)
    pred_pos = y_hat == cls
    if sum(pred_pos) > 0:
        return (y_hat[pred_pos] == y[pred_pos]).sum()/pred_pos.sum()
    else:
        return None


def recall(y_hat, y, cls):
    """
    Function to calculate the recall

    Inputs:
    > y_hat: pd.Series of predictions
    > y: pd.Series of ground truth
    > cls: The class chosen
    Output:
    > Returns the recall as float
    """
    assert(y_hat.size == y.size)
    assert(y_hat.size > 0)
    act_pos = y == cls
    if sum(act_pos) > 0:
        return (y_hat[act_pos] == y[act_pos]).sum()/act_pos.sum()
    else:
        return None


def confusion_mat(y_hat, y):
    assert(len(y_hat) == len(y))

    y_hat = y_hat > 0.5
    c = confusion_matrix(y, y_hat)
    print(c)