Foo

codes/partition/partition.py

@@ -4,6 +4,7 @@ from collections import defaultdict
 import hashlib
 import matplotlib.pyplot as plt
 import numpy as np
+import variance
 
 # int(ipaddress.IPv4Address("192.168.0.1"))
 
@@ -26,65 +27,74 @@ def partition(crawlers, ips):
 
 
 def main():
-    ips = load_ips('./ips.csv')
+    # ips = load_ips('./ips.csv')
+    ips = load_ips('./random.csv')
     c2 = ['c0', 'c1']
     c4 = ['c0', 'c1', 'c2', 'c3']
     c6 = ['c0', 'c1', 'c2', 'c3', 'c4', 'c5']
     c10 = ['c0', 'c1', 'c2', 'c3', 'c4', 'c5', 'c6', 'c7', 'c8', 'c9']
+    c100 = []
+    for i in range(100):
+        c100.append(f'c{i:02d}')
 
-    for cs in [c2, c4, c6, c10]:
+    # for cs in [c2, c4, c6, c10]:
+    for cs in [c100]:
         print(f'{len(cs)} crawlers')
 
         part = partition(cs, ips)
-
-        ind = np.arange(len(cs))
-        fig, ax = plt.subplots()
-        crawlers = sorted(cs)
-
-        ax.invert_yaxis()
-        ax.set_ylabel('Crawlers')
-        ax.set_xlabel('Peers')
-        ax.set_title(f'IP Partitioning for {len(cs)} crawlers')
-        ax.set_yticks(ind)
-
-        # ax.set_xticks(ind) #, labels=cs)
-
-        ns = []
-
-        for c, i in sorted(part.items(), key=lambda kv: kv[0]):
-            print(f'\t{c}: {len(i)}')
-            ns.append(len(i))
-
-        # p1 = ax.bar(ind, ns, 0.35)
-        hbars = ax.barh(ind, ns)
-        ax.bar_label(hbars, label_type='center')
-        # plt.show()
-        plt.savefig(f'./ip_part_c{len(cs):02}.png')
+        counts = [len(v) for k, v in part.items()]
+        variance.variance(counts)
+        # for c, i in sorted(part.items(), key=lambda kv: kv[0]):
 
 
-            # p1 = ax.bar(ind, i, 0.35, label=c)
+#         ind = np.arange(len(cs))
+#         fig, ax = plt.subplots()
+#         crawlers = sorted(cs)
+
+#         ax.invert_yaxis()
+#         ax.set_ylabel('Crawlers')
+#         ax.set_xlabel('Peers')
+#         ax.set_title(f'IP Partitioning for {len(cs)} crawlers')
+#         ax.set_yticks(ind)
+
+#         # ax.set_xticks(ind) #, labels=cs)
+
+#         ns = []
+
+#         for c, i in sorted(part.items(), key=lambda kv: kv[0]):
+#             print(f'\t{c}: {len(i)}')
+#             ns.append(len(i))
+
+#         # p1 = ax.bar(ind, ns, 0.35)
+#         hbars = ax.barh(ind, ns)
+#         ax.bar_label(hbars, label_type='center')
+#         # plt.show()
+#         plt.savefig(f'./ip_part_c{len(cs):02}.png')
 
 
-        # people = ('Tom', 'Dick', 'Harry', 'Slim', 'Jim')
+#             # p1 = ax.bar(ind, i, 0.35, label=c)
+
+
+#         # people = ('Tom', 'Dick', 'Harry', 'Slim', 'Jim')
 
 
 
-#         hbars = ax.barh(y_pos, performance, xerr=error, align='center')
-#         ax.set_yticks(y_pos, labels=people)
-#         ax.invert_yaxis()  # labels read top-to-bottom
-#         ax.set_xlabel('Performance')
-#         ax.set_title('How fast do you want to go today?')
+# #         hbars = ax.barh(y_pos, performance, xerr=error, align='center')
+# #         ax.set_yticks(y_pos, labels=people)
+# #         ax.invert_yaxis()  # labels read top-to-bottom
+# #         ax.set_xlabel('Performance')
+# #         ax.set_title('How fast do you want to go today?')
 
-#         # Label with specially formatted floats
-#         ax.bar_label(hbars, fmt='%.2f')
-#         ax.set_xlim(right=15)  # adjust xlim to fit labels
+# #         # Label with specially formatted floats
+# #         ax.bar_label(hbars, fmt='%.2f')
+# #         ax.set_xlim(right=15)  # adjust xlim to fit labels
 
-        # plt.show()
+#         # plt.show()
 
-        print()
-        print()
-        print()
-        print()
+#         print()
+#         print()
+#         print()
+#         print()
 
 
 if __name__ == '__main__':

codes/partition/variance.py

@@ -0,0 +1,51 @@
+#!/usr/bin/env python3
+
+from math import sqrt
+# from itertools import islice
+
+def variance(xs):
+    n = len(xs)
+    mu = sum(xs) / n
+    # mu = 265.8
+    # print(sum(map(lambda x: (x - mu)**2, xs)))
+    s = sum(map(lambda x: (x - mu)**2, xs))
+    var = s / n
+    deriv = sqrt(var)
+    # var = sqrt( / len(xs))
+
+    print('\\begin{align*}')
+    print(f'  n &= {n} \\\\')
+    print(f'  \\mu &= \\frac{{{sum(xs)}}}{{n}} = {mu:.1f} \\\\')
+    print(f'  s &= \\sum\\limits_{{i=1}}^{{n}} {{(x_i - \\mu)}}^2 \\\\')
+    # print [mylist[i:i+4] ]
+    first = True
+    for i in range(0, n, 3):
+        symbol = '=' if first else '+'
+        l = xs[i:i+3]
+        a = ' + '.join(map(lambda x: f'{{({x} - {mu:.1f})}}^2', l))
+        first = False
+        print(f'    &{symbol} {a} \\\\')
+
+    print(f'    &= {s:.1f} \\\\')
+
+    # a = ' + '.join(map(lambda x: f'({x} - {mu})^2', xs))
+    print(f'  \\sigma^2 &= \\frac{{s}}{{n}} = {var:.1f} \\\\')
+    print(f'  \\sigma &= \\sqrt{{\\sigma^2}} = {deriv:.1f}')
+    print('\\end{align*}')
+
+    print()
+    print()
+    print()
+    print()
+
+  # \mu &= 1595 / 4 = 398.75 \\
+  # \sigma &= \sqrt{\frac{(403 - 398.75)^2 + (369 - 398.75)^2 + (405 - 398.75)^2 + (418 - 398.75)^2}{4}} = 10.5
+
+def main():
+    # for xs in [[808, 787], [403, 369, 405, 418], [258, 273, 257, 264, 293, 250], [140, 175, 186, 166, 159, 152, 172, 148, 151, 146]]:
+    # for xs in [[258, 273, 257, 264, 293, 250]]:
+    for xs in [[499322, 500678], [249504, 250451, 249818, 250227], [166430, 166861, 166269, 166937, 166623, 166880], [100424, 99650, 99307, 100305, 99403, 100562, 100277, 99875, 99911, 100286]]:
+        variance(xs)
+
+if __name__ == '__main__':
+    main()