clustering.py

# encoding: utf-8
import csv
import json
import math

import numpy as np
import matplotlib.pyplot as plt
from scipy.spatial.distance import pdist, squareform
from sklearn.cluster import DBSCAN
from sklearn import metrics


X = []
with open("staypoints.csv","rb") as f:
    reader = csv.reader(f)
    for line in reader:
        X.append(line)
    X = np.array(X, np.float)


def calc_distance(lat1, lon1, lat2, lon2):
    theta = lon1 -lon2
    dist = math.sin(math.radians(lat1))*math.sin(math.radians(lat2)) \
            + math.cos(math.radians(lat1))*math.cos(math.radians(lat2))\
            * math.cos(math.radians(theta))
    if dist - 1 > 0 :
        dist = 1
    elif dist +1 < 0 :
        dist = -1
    dist = math.acos(dist)
    dist = math.degrees(dist)
    miles = dist * 60 * 1.1515
    return miles


def distance(p1, p2):
    lat1, lon1 = p1
    lat2, lon2 = p2

    distance = calc_distance(lat1, lon1, lat2, lon2) * 1.609344
    return distance


distance_matrix = squareform(pdist(X, (lambda u,v: distance(u,v))))
db = DBSCAN(eps=0.05, min_samples=3, metric='precomputed')
labels = db.fit_predict(distance_matrix)

# Plot
fig = plt.figure(1)
col = 'k'
plt.plot(X[:, 0], X[:, 1], '*', markerfacecolor='k', markeredgecolor='k', markersize=5)

core_samples_mask = np.zeros_like(labels, dtype=bool)
core_samples_mask[db.core_sample_indices_] = True

n_clusters_ = len(set(labels)) - (1 if -1 in labels else 0)
unique_labels = set(labels)
colors = plt.cm.Spectral(np.linspace(0, 1, len(unique_labels)))
fig = plt.figure(2)
center_point = []
heatmap_point = []

for k, col in zip(unique_labels, colors):
    if k == -1:
        col = 'k'

    class_member_mask = (labels == k)
    xy = X[class_member_mask & core_samples_mask]

    print "%d reference points contain %d points" %(k,len(xy))
    plt.plot(xy[:, 0], xy[:, 1], 'o', markerfacecolor=col,
             markeredgecolor='k', markersize=14)

    print "center pos %f %f" %(np.mean(xy[:, 0]), np.mean(xy[:, 1]) )
    center_point.append([np.mean(xy[:, 1]), np.mean(xy[:, 0])])
    heatmap_point.append({"lng": np.mean(xy[:, 1]), "lat": np.mean(xy[:, 0]), "count": len(xy[:, 1])})


# json data for baidu map
with open('clustering_points.js', 'w') as f:
    f.write('var data = {"data": %s}' % json.dumps(center_point))
with open('heatmap_points.js', 'w') as f:
    f.write('var points = %s' % json.dumps(heatmap_point))


plt.title('DBSCAN :Estimated number of clusters: %d' % n_clusters_)
print("Silhouette Coefficient: %0.3f" % metrics.silhouette_score(X, labels))
plt.show()
print "Done."