Задание по программированию: Размещение баннеров

In [1]:

    

import pandas as pd


    

In [2]:

    

data = pd.read_csv('preprocessed_data.csv')

data['created_at'] = pd.to_datetime(data['created_at'])


    

In [3]:

    

data.head()


    

    
Out[3]:







  

    

      

      
id
      
user_id
      
venue_id
      
latitude
      
longitude
      
created_at
    
  
  

    

      
0
      
984222
      
15824
      
5222
      
38.895112
      
-77.036366
      
2012-04-21 17:43:47
    
    

      
1
      
984234
      
44652
      
5222
      
33.800745
      
-84.410520
      
2012-04-21 17:43:43
    
    

      
2
      
984291
      
105054
      
5222
      
45.523452
      
-122.676207
      
2012-04-21 17:39:22
    
    

      
3
      
984318
      
2146539
      
5222
      
40.764462
      
-111.904565
      
2012-04-21 17:35:46
    
    

      
4
      
984232
      
93870
      
380645
      
33.448377
      
-112.074037
      
2012-04-21 17:38:18
    
  

In [4]:

    

subset = data.values[0:100000, 3:5]


    

In [5]:

    

import numpy as np
from sklearn.cluster import MeanShift, estimate_bandwidth


    

In [6]:

    

ms = MeanShift(bandwidth=0.1)
ms.fit(subset)

labels = ms.labels_
cluster_centers = ms.cluster_centers_

labels_unique = np.unique(labels)
n_clusters_ = len(labels_unique)

print("number of estimated clusters : %d" % n_clusters_)


    

    




number of estimated clusters : 3230

In [7]:

    

import matplotlib.pyplot as plt
from itertools import cycle

plt.figure(1)
plt.clf()

colors = cycle('bgrcmykbgrcmykbgrcmykbgrcmyk')
for k, col in zip(range(n_clusters_), colors):
    my_members = labels == k
    cluster_center = cluster_centers[k]
    plt.plot(subset[my_members, 0], subset[my_members, 1], col + '.')
    plt.plot(cluster_center[0], cluster_center[1], 'o', markerfacecolor=col,
             markeredgecolor='k', markersize=14)
plt.title('Estimated number of clusters: %d' % n_clusters_)
plt.show()


    

In [8]:

    

clusters_dict = dict()
for label in labels:
    if label not in clusters_dict.keys():
        clusters_dict[label]  = 1
    else:
        clusters_dict[label] += 1

count = 0
for key in clusters_dict.keys():
    if clusters_dict[key] > 15:
        count += 1

clusters_select = np.ndarray(shape=(count, 2))

j = 0
for i in range(len(cluster_centers)):
    if clusters_dict[i] > 15:
        clusters_select[j] = cluster_centers[i]
        j += 1


    

In [9]:

    

offices = np.ndarray(shape=(6,2))
offices[0] = np.array([33.751277, -118.188740])
offices[1] = np.array([25.867736, -80.324116])
offices[2] = np.array([51.503016, -0.075479])
offices[3] = np.array([52.378894, 4.885084])
offices[4] = np.array([39.366487, 117.036146])
offices[5] = np.array([-33.868457, 151.205134])


    

In [10]:

    

def distance(point1, point2):
    return ((point1[0] - point2[0])**2 + (point1[1] - point2[1])**2)**0.5


    

In [11]:

    

answer_index = 0
min_dist = 0
for i in range(len(clusters_select)):
    distances = [distance(xx, clusters_select[i]) for xx in offices]
    if min_dist == 0:
        min_dist = min(distances)
        answer_index = i
    elif min_dist > min(distances):
        min_dist = min(distances)
        answer_index = i


    

In [12]:

    

def write_answer(center):
    with open("answer.txt", "w") as f:
        f.write(str(center[0]) + ' ' + str(center[1]))


    

In [13]:

    

write_answer(clusters_select[answer_index])