/
NYenergy.py
373 lines (316 loc) · 15.5 KB
/
NYenergy.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
import time, sys, os, csv, random, json,sys, logging
import matplotlib
import numpy as np
import pylab as plt
from scipy.stats.stats import pearsonr
from sklearn import svm, metrics, preprocessing
from sklearn import cross_validation
from imposm.parser import OSMParser
import citygraph
cityname = "NewYork"
original_pbf = "/Users/vallette/projects/DATA/citygraph/pbf/new-york.osm.pbf"
families = {
"bank":{"amenity":["bank", "atm", "bureau_de_change"]},
"entertainment":{"amenity":["cinema","community_center","nightclub","stripclub","theatre"],
"historic":"*",
"leisure":"*",
"sport":"*",
"tourism":"*",
},
"storage":{
"building":["industrial","warehouse"],
},
"health":{
"amenity":["baby_hatch","clinic","dentist","doctors","hospital","social_facility"]
},
"education":{
"amenity":["colledge","kindergarten","library","school","university"]
},
"hotel":{
"building":["hotel",]
},
"living":{
"building":["apartments","dormitory","house","residential"]
},
"commercial":{"shop":"*",
"amenity":["car_rental","marketplace"],
"building":["commercial","retail"],
"landuse":["commercial",]
},
"office":{"office":"*",
}
}
logger = logging.getLogger('root')
formatter = logging.Formatter(fmt='%(asctime)s - %(levelname)s - %(module)s - %(message)s')
handler = logging.FileHandler("/Users/vallette/projects/DATA/epml/outputs/log_%s.log" % cityname)
handler.setFormatter(formatter)
logger.addHandler(handler)
class Building(citygraph.Feature):
def __init__(self, id, lon, lat, btype, conso):
self.bbl = id
self.btype = btype
self.conso = conso
if conso != None:
super(Building, self).__init__(lon, lat, "monitored_building", "building")
else:
super(Building, self).__init__(lon, lat, "unmonitored_building", "building")
class ElecSection(object):
def __init__(self, section, input):
self.section = section
self.input = input
self.result = False
class NewCity(citygraph.City):
def __init__(self, name, top, left, bottom, right):
super(NewCity, self).__init__(name, top, left, bottom, right)
self.classifier = None
self.elecsections = []
def load_osm_features(self, pbf_path , families):
logger.info(" Adding feature from OSM.")
pbf_path, pbf_name = os.path.split(pbf_path)
categories = {}
for family in families:
locations = {}
class FamilyParser(object):
ways_objects = {}
nodes_objects = {}
all_nodes = {}
def ways(self, ways):
for osmid, tags, refs in ways:
for key in families[family].keys():
if key in tags.keys():
if families[family][key] == "*" or tags[key] in families[family][key]:
self.ways_objects[int(osmid)] = {"nd" : refs, "tag" : tags[key]}
def nodes(self, nodes):
for osmid, tags, coords in nodes:
for key in families[family].keys():
if key in tags.keys():
if families[family][key] == "*" or tags[key] in families[family][key]:
self.nodes_objects[int(osmid)] = {"lon" : coords[0], "lat" : coords[1], "tag" : tags[key]}
def coords(self, coords):
for coords in coords:
self.all_nodes[int(coords[0])] = (coords[1], coords[2])
parser = FamilyParser()
p = OSMParser(concurrency = 2, ways_callback = parser.ways, nodes_callback = parser.nodes, coords_callback = parser.coords)
p.parse("%s/%s.osm.pbf" % (pbf_path, self.name))
for k in parser.nodes_objects.keys():
locations[k] = (parser.nodes_objects[k]["lon"], parser.nodes_objects[k]["lat"], parser.nodes_objects[k]["tag"])
for k in parser.ways_objects.keys():
# compute mean postion
m_lon, m_lat = 0, 0
for nd in parser.ways_objects[k]["nd"]:
node = parser.all_nodes[int(nd)]
m_lon += node[0]
m_lat += node[1]
m_lon /= float(len(parser.ways_objects[k]["nd"]))
m_lat /= float(len(parser.ways_objects[k]["nd"]))
locations[nd] = (m_lon, m_lat, parser.ways_objects[k]["tag"]) #a way can have the same id as a node so we take the id of the last node
logger.info("Number of elements in family %s : %s" % (family, str(len(locations.keys()))))
categories[family] = locations
for category in categories.keys():
for id in categories[category].keys():
building = Building(int(id), categories[category][id][0], categories[category][id][1], category, None)
self.features += [building]
def load_electricity_consumption(self):
category_map = {
'': "unknown",
'Other': "unknown",
'Data Center': "unknown",
'Bank/Financial Institution': "bank",
'Entertainment/Culture': "entertainment",
'Self-Storage': "storage",
'Health Care: Outpatient': "health",
'College/University (Campus-Level)': "education",
'Senior Care Facility': "health",
'Hotel': "hotel",
'Multifamily Housing': "living",
'Hospital (General Medical and Surgical)': "health",
'Recreation': "entertainment",
'Medical Office': "health",
'Warehouse (Unrefrigerated)': "storage",
'Residence Hall/Dormitory': "living",
'Retail': "commercial",
'Storage/Shipping/Non-Refrigerated Warehouse': "storage",
'Office': "office",
'Retail (Misc)': "commercial",
'Service (Vehicle Repair/Service, Postal Service)': "commercial",
'Clinic/Other Outpatient Health': "health",
'Lodging': "living",
'K-12 School': "education",
'Education': "education",
'House of Worship': "entertainment",
"Health Care: Inpatient (Specialty Hospitals, Excluding Children's)": "health",
'Automobile Dealership': "commercial",
'Public Assembly': "entertainment",
'Social/Meeting': "entertainment",
'Warehouse (Refrigerated)': "storage",
'Supermarket/Grocery': "commercial"
}
logger.info(" Adding features from Nonresidential Benchmark Data-Tableau 1")
with open("/Users/vallette/projects/DATA/epml/csv/Nonresidential_Benchmark_Data-Tableau_1.csv", "rb") as csvfile:
spamreader = csv.reader(csvfile, delimiter=";")
next(spamreader) # skip first line
geofile = open("/Users/vallette/projects/DATA/epml/json/Nonresidential_Benchmark_Data-Tableau_geocoded.json", "r")
geocoded = json.loads(geofile.read())
n_id = 0
types = []
for row in spamreader:
bbl = int(row[0])
house_num = row[1]
street = row[2]
district = row[3]
if geocoded.has_key(str(bbl)):
lon, lat = geocoded[str(bbl)]
if self.point_inside_polygon(lon, lat):
try:
conso = float(row[10].replace(",","."))
btype = row[13]
types += [btype]
if category_map.has_key(btype):
if category_map[btype] != "unknown":
building = Building(bbl, lon, lat, category_map[btype], conso)
self.features += [building]
n_id += 1
else:
print "Missing mapping for %s" % btype
f=lambda s,d={}:([d.__setitem__(i,d.get(i,0)+1) for i in s],d)[-1]
print f(types)
sys.exit(0)
except:
pass
logger.info(" %s buildings with conso" % str(n_id))
def create_ElecSections(self):
self.elecsections = []
maximums = np.zeros(13)
minimums = np.zeros(13)
resmax = 0
resmin = 0
for section in self.sections.values():
x,y = self.map.map(np.mean([n.lon for n in section.nodes]), np.mean([n.lat for n in section.nodes]))
type_dict = {'living_street' : 1.,'residential' : 2.,'primary': 6.,'unclassified': 3.,'tertiary' : 4.,'secondary': 5.}
section_type = type_dict[section.type]
list = [x, y, section.Vmax, section_type]
cat_index = {"bank":0, "entertainment":1, "storage":2, "health":3, "education":4, "hotel":5, "living":6, "commercial":7, "office":8}
tab = np.zeros(9)
for building in section.features:
tab[cat_index[building.btype]] += 1
if sum(tab) > 0:
input_vec = np.append(np.array(list), tab)
elecsection = ElecSection(section, input_vec)
maximums = np.maximum(maximums, input_vec)
minimums = np.minimum(minimums, input_vec)
buildings = [f for f in section.features if f.tag == "monitored_building"]
if len(buildings) > 0:
mean_conso = sum([float(b.conso) for b in buildings])
resmax = np.max([resmax, mean_conso])
resmin = np.min([resmin, mean_conso])
elecsection.result = float(mean_conso)
self.elecsections += [elecsection]
# normalization
for elecsection in self.elecsections:
elecsection.input = (elecsection.input-minimums)/(maximums-minimums)
if elecsection.result:
elecsection.result = (elecsection.result - resmin)/(resmax - resmin)
def learn(self):
#make data
data = np.array([], dtype="float64", order="C")
result = np.array([], dtype="float64", order="C")
for elecsection in self.elecsections:
if elecsection.result:
result = np.append(result, elecsection.result)
if len(data) > 0:
data = np.vstack((data, elecsection.input))
else:
data = np.array(elecsection.input)
n_samples = len(result)
logger.info("%s samples" % str(n_samples))
indexer = np.arange(n_samples)
random.shuffle(indexer)
result_s = result[indexer]
data_s = data[indexer]
# learn
size = n_samples/2
self.classifier = svm.NuSVR()
self.classifier.fit(data_s[:size], result_s[:size])
expected = result_s[size:]
predicted = self.classifier.predict(data_s[size:])
logger.info(" Pearson coefficient: %s" % str(pearsonr(expected,predicted)[0]))
scores = cross_validation.cross_val_score(self.classifier, data_s, result_s, cv=5)
logger.info("Accuracy: %0.2f (+/- %0.2f)" % (scores.mean(), scores.std() / 2))
def plot(self, list):
fig = plt.figure(figsize=(9, 9))
ax = fig.add_subplot(111)
if "highways" in list:
for highway in self.highways:
lons = [node.lon for node in highway.nodes]
lats = [node.lat for node in highway.nodes]
x, y = self.map.map(lons, lats)
ax.plot(x, y, "-", linewidth = 0.5, color = "k")
if "buildings" in list:
for feature in self.features:
x, y = self.map.map(feature.lon, feature.lat)
if feature.tag == "monitored_building":
ax.scatter(x, y, marker = "o", color = "r", s = 3)
else:
ax.scatter(x, y, marker = "o", color = "b", s = 2)
if "heat" in list:
xall, yall, call = [], [], []
cm = plt.get_cmap('YlOrRd')
cNorm = matplotlib.colors.Normalize(vmin=0, vmax=0.1)
scalarMap = matplotlib.cm.ScalarMappable(norm=cNorm, cmap=cm)
for elecsection in self.elecsections:
section = elecsection.section
lons = [node.lon for node in section.nodes]
lats = [node.lat for node in section.nodes]
x, y = self.map.map(lons, lats)
colorVal = scalarMap.to_rgba(self.classifier.predict(elecsection.input)[0])
ax.plot(x, y, "-", linewidth = 5, color=colorVal, alpha=0.5)
ax.axis("equal")
plt.xticks([])
plt.yticks([])
if "pdf" in list:
fig = plt.gcf()
fig.set_size_inches(20,20)
plt.savefig("fig.pdf")
else:
plt.show()
def feature_distribution(self):
hist_families = {"bank":[], "entertainment":[], "storage":[], "health":[], "education":[], "hotel":[], "living":[], "commercial":[], "office":[]}
for building in self.features:
hist_families[building.btype] += [building.conso]
importances = np.array([np.mean(hist_families[family]) for family in hist_families.keys()])
std = np.array([np.std(hist_families[family]) for family in hist_families.keys()])
indices = np.argsort(importances)[::-1]
fig = plt.figure()
plt.title("Feature importances")
ax = fig.add_subplot(111)
ax.bar(xrange(len(hist_families.keys())), importances[indices], color="r", yerr=std[indices], align="center")
plt.xticks(xrange(len(hist_families.keys())), [hist_families.keys()[i] for i in indices],rotation=90)
plt.xlim([-1, len(hist_families.keys())])
plt.show()
NY = NewCity(cityname, top=41.162, left=-74.593, bottom=40.229, right=-73.365)
NY.load_highways_from_osm(original_pbf, [189005738, 189020941, 46936809, 37788309, 37788311, 46945175, 37788079, 37788080, 37788405, 37788412, 37788493, 37811053, 37787073,
37787072, 37787122, 37811067, 37787976, 37787975, 37787971, 37788522, 61603751, 61928627, 61922976, 61603744, 61923037, 61909165, 37565950,
37565866, 37565868, 37565564, 37565158, 37565053, 37565052, 61603753, 37565051, 37569166, 61602967, 61602968, 61602961, 61602961, 61602969,
61602963, 61602964, 61602965, 37567463, 37567690, 189026085, 37567199, 37567197], reverse = True, redo_pbf = False)
NY.make_sections()
NY.compute_connectivity()
NY.clean_sections(plot_check = False)
NY.load_graph()
NY.load_osm_features(original_pbf, families)
NY.load_electricity_consumption()
NY.associate_features(precision = 5)
NY.create_ElecSections()
NY.learn()
# import json
# out = []
# for elecsection in NY.elecsections:
# section = elecsection.section
# lons = [node.lon for node in section.nodes]
# lats = [node.lat for node in section.nodes]
# lonlat = [[lon,lat] for lon, lat in zip(lons, lats)]
# val = NY.classifier.predict(elecsection.input)[0]
# s = {"coords" : lonlat, "id" : section.id, "value" : val}
# out += [s]
# outfile = open("empl.json", "w")
# outfile.write(json.dumps(out))
# outfile.close()