Code review!

MaxFlowUtils.py

@@ -1,22 +1,25 @@
 import math, collections
 from MaxFlowLib import FlowEdge
-import matplotlib.pyplot as plt
+# import matplotlib.pyplot as plt
 import numpy as np
 import copy
 _DISTANCE = .4 	# km
 
 # vectorized implementation
 def distance(v, stations):
-	'''find distance from v to all other vertices'''
-	R = 6371		
-	dLat_dLong = np.radians(stations[:, [0, 1]] - stations[v, [0, 1]]) 
-	dLat = dLat_dLong[:, 0]
-	dLong = dLat_dLong[:, 1]
-	lat = np.radians(stations[:, 0])  		# col vector of lat values for all stations; num_stations x 1
-	lat_v = np.radians(stations[v][0])    	# scalar for this station
-
-	dLatH = np.divide(dLat, 2)
-	dLongH =  np.divide(dLong, 2)
+	'''
+	find distance from v to all other vertices
+	input: stations array of [(lat, long), (lat, long) ...] 
+	output: distance array of [dist, dist, ... ]
+	'''
+	stations = np.radians(stations)
+	R = 6371 # ?
+	dLat_dLong = stations - stations[v]
+	dLatH = dLat_dLong[:, 0] / 2
+	dLongH = dLat_dLong[:, 1] / 2
+	lat = stations[:, 0]		# col vector of lat values for all stations; num_stations x 1
+	lat_v = stations[v][0]    	# scalar for this station
+
 	a = np.sin(dLatH) ** 2 + np.sin(dLongH) ** 2 * np.cos(lat) * np.cos(lat_v)
 	c = np.arctan2(np.sqrt(a), np.sqrt(1-a)) * 2
 	d = c * R
@@ -32,8 +35,9 @@ def findFlowEdges(stations, totalDocks, source, capacity_func):
 	'''
 	flow_edges = collections.deque()
 	num_stations = len(stations)
-	marked_v = np.ones((num_stations,), dtype=bool) # marks all from vertices
-	marked_w = np.ones((num_stations,), dtype=bool) # marks all to vertices
+	marked_v = np.ones(num_stations, dtype=bool) # marks all from vertices
+	marked_w = np.ones(num_stations, dtype=bool) # marks all to vertices
+	# this feels like there should be an object
 	q = collections.deque()
 	q.append(source)
 

NetworkFlowAnalyzer.py

@@ -11,7 +11,7 @@
 '''
 import json
 import argparse
-import matplotlib.pyplot as plt
+# import matplotlib.pyplot as plt
 import numpy as np
 import random
 import MaxFlowUtils
@@ -34,22 +34,20 @@
     input = json.load(f)
 
 num_stations = len(input["stationBeanList"])
-stations = np.random.random((num_stations, 2))  # 2 cols for latitude, longitude
+stations = np.zeros(shape=(num_stations, 2))  # 2 cols for latitude, longitude
 #totalDocks = np.zeros((num_stations,), dtype=np.uint8)
 totalDocks = np.zeros((num_stations,), dtype=np.int16)
 names = {}  # maps vertex number to stationNam
 vertex = {} # maps stationName to vertex number
 
 # store data
-row = 0
-for station in input["stationBeanList"]:
+for row, station in enumerate(input["stationBeanList"]):
 	# store data in numpy arrays
     totalDocks[row] = station["totalDocks"]
     stations[row][0] = station["latitude"]
     stations[row][1] = station["longitude"]
     vertex[station["stationName"]] = row
     names[row] = station["stationName"]
-    row += 1
 
 # process optional args
 if args.stations:
@@ -78,15 +76,15 @@
         # calculate maxflow using Ford-Fulkerson algorithm
         maxflow = FF(flownet, source, target)
         # plot flow network
-        MaxFlowUtils.plotFlowNetwork(stations, source, target, flow_edges)
-        plt.show()
+        # MaxFlowUtils.plotFlowNetwork(stations, source, target, flow_edges)
+        # plt.show()
         # flow_edges.append(FlowEdge(329, 330, totalDocks[329] + totalDocks[330]))
 
         # plot flow
         flowpath = MaxFlowUtils.flowPath(stations, flownet, source)
         flowpath2 = copy.copy(flowpath)
-        MaxFlowUtils.plotFlow(stations, source, target, flowpath)
-        plt.show()
+        # MaxFlowUtils.plotFlow(stations, source, target, flowpath)
+        # plt.show()
 
         # convert vertices in flow path to station names
         station_path = MaxFlowUtils.toStationNames(flowpath2, names)
@@ -105,8 +103,8 @@
             print e
 
         # plot edges in st-cut
-        MaxFlowUtils.plotSTcut(stations, source, target, flowpath, stcut_v)
-        plt.show()
+        # MaxFlowUtils.plotSTcut(stations, source, target, flowpath, stcut_v)
+        # plt.show()
     else:
         print '{} and {} are not connected.'.format(start_station, end_station)
 

README

@@ -10,7 +10,7 @@ Capacity on an edge is the max of the bike docks at each edge vertex.
 usage: run NetworkFlowAnalyzer.py "citybike.json" <start-station> <end-station>
 
 Example:
-run NetworkFlowAnalyzer.py "citybike.json" "Broadway & Battery Pl" "E 47 St & 1 Ave"
+run NetworkFlowAnalyzer.py --start_station "Broadway & Battery Pl" --end_station "E 47 St & 1 Ave" "citybike.json"
 
 run NetworkFlowAnalyzer.py [-h] [--stations] [--start_station START_STATION] [--end_station END_STATION] filename-json
 default start_station: "Broadway & Battery Pl"