"""
Felix Muzny
November 3rd/4th, 2021
DS 2001 - CS Practicum #8

The first part of implementing k-means clustering
for 2-dimensional points!
"""


import kmeans_utils as utils
import matplotlib.pyplot as plt
import math

# Task 4
def distance(p1, p2):
    """
    name: distance
    This function takes two 2-d points and calculates the Euclidian distance between them.
    parameters:
    point1 - list of two numbers
    point2 - list of two numbers
    return:
    float euclidian distance between the two points (x1, y1) and (x2, y2)
    """
    d = math.sqrt((p1[0] - p2[0]) ** 2 + (p1[1] - p2[1]) ** 2)
    return d 
    
# Task 5
def closest(target_point, point_list):
    """
    name: closest
    This function takes a target point and a list of points
    and calculates the Euclidian distance between them
    to find the point in the list that is closest to the target.
    parameters:
    target_point - list of a 2d point
    point_list - list of lists of 2d points
    return:
    the tuple point from point_list that is closest to target_point
    """
    closest = None
    current_min = -1
    for point in point_list:
        d = distance(target_point, point)
        if d < current_min or closest is None:
            closest = point
            current_min = d
    return closest

if __name__ == "__main__":
    # Task 1
    animal_ratings = {"cat": 4, "dog":4, "whale": 100}
    print(animal_ratings)

    # Task 2
    points = utils.generate_points(20)
    centroids = utils.generate_points(2)
    print("data:", points)
    print("centroids:", centroids)
    print()

    # graphing our data
    plt.scatter(utils.get_column(points, 0), utils.get_column(points, 1), label="data")
    plt.scatter(utils.get_column(centroids, 0), utils.get_column(centroids, 1), label="centroids")
    plt.xlabel("x value")
    plt.ylabel("y value")
    plt.title("initial data for k-means clustering")
    plt.legend()
    plt.show()

    # Testing task 3
    print("Testing distance function")
    print("distance from (0, 0) to (1, 1)", distance((0, 0), (1, 1)))
    print("distance from (-10, 0) to (10, 0)", distance((-10, 0), (10, 0)))

    print("first point in the data list:", points[0])
    print("first point in the centroids list:", centroids[0])
    print("distance between them:", distance(points[0], centroids[0]))

    print()
    target = [1, 1]

    # Testing task 4
    print("Testing closest function")
    print("closest centroid to (0, 0):", closest((0, 0), centroids))
    print("closest centroid to (10, 10):", closest((10, 10), centroids))
    print()

    # Task 5
    clusters = {}
    for point in points:
        belongs_to = closest(point, centroids)
        if belongs_to not in clusters:
            clusters[belongs_to] = 0
        clusters[belongs_to] += 1

    print("counts of points assigned to clusters")
    print(clusters)