-
Notifications
You must be signed in to change notification settings - Fork 0
/
classical_embed_corner_bills.py
319 lines (275 loc) · 11.1 KB
/
classical_embed_corner_bills.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
import csv
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.lines import Line2D
from mpl_toolkits.mplot3d import Axes3D
import sys
import math
if len(sys.argv) >= 2:
congress = sys.argv[1]
else:
congress = 'H100'
r_list = [0.05, 0.1, 0.15, 0.2, 0.25]
f_ll_list = []
f_lr_list = []
f_ul_list = []
f_ur_list = []
c_list = []
r = 0.2
f_ll_bills = []
f_lr_bills = []
f_ul_bills = []
f_ur_bills = []
f_bills = []
total = 0
tokens = dict()
for i in range(115, 116):
print('Looking at congress '+str(i))
if i <= 99:
congress = 'H0'+str(i)
else:
congress = 'H'+str(i)
data_folder = 'output/raw/'+congress+'/'+congress+'_'
output_folder = 'output/raw/'+congress+'/'+congress+'_'
write_folder = 'output/corner_analysis/'
hd_file = 'output/house_details.csv'
file_name = congress+'_corner_bills'
bill_mat = np.genfromtxt(output_folder+'eigenbills_squared_normalized.csv', delimiter=',')
bill_mat = bill_mat[:,2:]
bill_details = []
with open(output_folder+'eigenbills.csv') as csvfile:
csv_reader = csv.reader(csvfile, delimiter=',')
for row in csv_reader:
bill_details.append(row)
for token in bill_details[-1][-1].split(' '):
if token.lower() not in tokens:
tokens[token.lower()] = 1
else:
tokens[token.lower()] = tokens[token.lower()]+1
x_min = np.min(bill_mat[:,0])
x_max = np.max(bill_mat[:,0])
y_min = np.min(bill_mat[:,1])
y_max = np.max(bill_mat[:,1])
mask1 = (bill_mat[:, 0] <= x_min+r*(x_max-x_min))
mask2 = (bill_mat[:, 1] <= y_min+r*(y_max-y_min))
f_ll = []
mask = mask1*mask2
for i in range(mask.shape[0]):
if mask[i] == True:
f_ll.append(i)
mask1 = (bill_mat[:, 0] >= x_max-r*(x_max-x_min))
mask2 = (bill_mat[:, 1] <= y_min+r*(y_max-y_min))
f_lr = []
mask = mask1*mask2
for i in range(mask.shape[0]):
if mask[i] == True:
f_lr.append(i)
mask1 = (bill_mat[:, 0] <= x_min+r*(x_max-x_min))
mask2 = (bill_mat[:, 1] >= y_max-r*(y_max-y_min))
f_ul = []
mask = mask1*mask2
for i in range(mask.shape[0]):
if mask[i] == True:
f_ul.append(i)
mask1 = (bill_mat[:, 0] >= x_max-r*(x_max-x_min))
mask2 = (bill_mat[:, 1] >= y_max-r*(y_max-y_min))
f_ur = []
mask = mask1*mask2
for i in range(mask.shape[0]):
if mask[i] == True:
f_ur.append(i)
for i in f_ll:
f_ll_bills.append(bill_details[i][-1])
for i in f_lr:
f_lr_bills.append(bill_details[i][-1])
for i in f_ul:
f_ul_bills.append(bill_details[i][-1])
for i in f_ur:
f_ur_bills.append(bill_details[i][-1])
for i in range(len(bill_details)):
if i not in f_ll and i not in f_lr and i not in f_ul and i not in f_ur:
f_bills.append(bill_details[i][-1])
total = total + len(bill_details)
print('Number of Tokens collected:')
#print(tokens.keys())
print(len(tokens.keys()))
print('Total bills:')
print(len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills)+len(f_ur_bills)+len(f_bills))
print(len(f_ll_bills))
print(len(f_lr_bills))
print(len(f_ul_bills))
print(len(f_ur_bills))
print(len(f_bills))
print(total)
def vectorize(sent_list):
vec_list = []
for sent in sent_list:
vec = np.zeros((len(tokens.keys()),1))
vec_dict = dict()
for token in tokens.keys():
vec_dict[token] = 0
for token in sent.split(' '):
vec_dict[token.lower()] = vec_dict[token.lower()]+1
i = 0
for token in tokens.keys():
vec[i] = vec_dict[token]
i = i + 1
vec = vec/sum(vec)
vec_list.append(vec)
return vec_list
import operator
sorted_tokens = sorted(tokens.items(), key=operator.itemgetter(1), reverse=True)
import copy
for i in range(1):
stopwords = [word[0] for word in sorted_tokens[:i]]
print(stopwords)
f_ll_b = copy.deepcopy(f_ll_bills)
f_lr_b = copy.deepcopy(f_lr_bills)
f_ul_b = copy.deepcopy(f_ul_bills)
f_ur_b = copy.deepcopy(f_ur_bills)
f_b = copy.deepcopy(f_bills)
all_b = f_ll_b+f_lr_b+f_ul_b+f_ur_b+f_b
for i in range(len(all_b)):
#print(all_b[i])
querywords = all_b[i].split()
resultwords = [word for word in querywords if word.lower() not in stopwords]
result = ' '.join(resultwords)
all_b[i] = result
#print(all_b[i])
vectorized = vectorize(all_b)
print('Done Running Vectorization')
if len(f_ll_bills) == 0:
vectors_ll = []
else:
vectors_ll = vectorized[0:len(f_ll_bills)]
if len(f_lr_bills) == 0:
vectors_lr = []
else:
vectors_lr = vectorized[len(f_ll_bills):len(f_ll_bills)+len(f_lr_bills)]
if len(f_ul_bills) == 0:
vectors_ul = []
else:
vectors_ul = vectorized[len(f_ll_bills)+len(f_lr_bills):len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills)]
if len(f_ur_bills) == 0:
vectors_ur = []
else:
vectors_ur = vectorized[len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills):len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills)+len(f_ur_bills)]
if len(f_bills) == 0:
vectors = []
else:
vectors = vectorized[len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills)+len(f_ur_bills):len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills)+len(f_ur_bills)+len(f_bills)]
from scipy import spatial
from sklearn.manifold import TSNE
from matplotlib import pyplot as plt
import matplotlib.patches as mpatches
vectors_corner = vectors_ll + vectors_lr + vectors_ul + vectors_ur
vectors_full = vectors_corner + vectors
vectors_full = np.stack(vectors_full).squeeze()
print(vectors_full.shape)
embedding = TSNE(metric='cosine')
embed_vectors = embedding.fit_transform(vectors_full)
print(embed_vectors.shape)
'''
colors = np.zeros((vectors_full.shape[0],))
for i in range(vectors_full.shape[0]):
if i < len(f_ll_bills):
colors[i] = 0.2
elif i < len(f_ll_bills)+len(f_lr_bills):
colors[i] = 0.4
elif i < len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills):
colors[i] = 0.6
elif i < len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills)+len(f_ur_bills):
colors[i] = 0.8
elif i < len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills)+len(f_ur_bills)+len(f_bills):
colors[i] = 1
cmap = plt.get_cmap('viridis')
print(cmap)
ll = mpatches.Patch(color=cmap[0.4], label='LL corner')
lr = mpatches.Patch(color=0.55, label='LR corner')
ul = mpatches.Patch(color=0.7, label='UL corner')
ur = mpatches.Patch(color=0.85, label='UR corner')
nc = mpatches.Patch(color=1, label='non-corner')
plt.legend(handles=[ll,lr,ul,ur,nc])
#ax.scatter(embed_vectors[:len(f_ll_bills),0], embed_vectors[:len(f_ll_bills),1], c=colors[:len(f_ll_bills)], label='LL corner')
#ax.scatter(embed_vectors[len(f_ll_bills):len(f_ll_bills)+len(f_lr_bills),0], embed_vectors[len(f_ll_bills):len(f_ll_bills)+len(f_lr_bills),1], c=colors[len(f_ll_bills):len(f_ll_bills)+len(f_lr_bills)], label='LR corner')
'''
fig, ax = plt.subplots()
ax.scatter(embed_vectors[:len(f_ll_bills),0], embed_vectors[:len(f_ll_bills),1], c='r', label='LL corner')
ax.scatter(embed_vectors[len(f_ll_bills):len(f_ll_bills)+len(f_lr_bills),0], embed_vectors[len(f_ll_bills):len(f_ll_bills)+len(f_lr_bills),1], c='b', label='LR corner')
ax.scatter(embed_vectors[len(f_ll_bills)+len(f_lr_bills):len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills),0], embed_vectors[len(f_ll_bills)+len(f_lr_bills):len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills),1], c='g', label='UL corner')
ax.scatter(embed_vectors[len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills):len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills)+len(f_ur_bills),0], embed_vectors[len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills):len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills)+len(f_ur_bills),1], c='y', label='UR corner')
ax.scatter(embed_vectors[len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills)+len(f_ur_bills):,0], embed_vectors[len(f_ll_bills)+len(f_lr_bills)+len(f_ul_bills)+len(f_ur_bills):,1], c='k', label='non-corner')
ax.legend()
plt.show()
count = 0
avg1 = 0.0
avg = 0.0
for i in range(len(vectors_ll)):
for j in range(i+1, len(vectors_ll)):
count = count + 1
avg = (count-1)*avg/count + spatial.distance.cosine(vectors_ll[i], vectors_ll[j])/count
avg1 = avg
count = 0
avg2 = 0
avg = 0
for i in range(len(vectors)):
for j in range(len(vectors_ll)):
count = count + 1
avg = (count-1)*avg/count + spatial.distance.cosine(vectors[i], vectors_ll[j])/count
avg2 = avg
print('Avg dist b/w ll corner bills, Avg dist b/w ll corner and non-corner bills : ' + str(round(avg1,2)) + ',' + str(round(avg2, 2)))
print('Percentage increase : ' + str(round(((avg2-avg1)/avg1),2)))
count = 0
avg1 = 0.0
avg = 0.0
for i in range(len(vectors_lr)):
for j in range(i+1, len(vectors_lr)):
count = count + 1
avg = (count-1)*avg/count + spatial.distance.cosine(vectors_lr[i], vectors_lr[j])/count
avg1 = avg
count = 0
avg2 = 0
avg = 0
for i in range(len(vectors)):
for j in range(len(vectors_lr)):
count = count + 1
avg = (count-1)*avg/count + spatial.distance.cosine(vectors[i], vectors_lr[j])/count
avg2 = avg
print('Avg dist b/w lr corner bills, Avg dist b/w lr corner and non-corner bills : ' + str(round(avg1,2)) + ',' + str(round(avg2, 2)))
print('Percentage increase : ' + str(round(((avg2-avg1)/avg1),2)))
count = 0
avg1 = 0.0
avg = 0.0
for i in range(len(vectors_ul)):
for j in range(i+1, len(vectors_ul)):
count = count + 1
avg = (count-1)*avg/count + spatial.distance.cosine(vectors_ul[i], vectors_ul[j])/count
avg1 = avg
count = 0
avg2 = 0
avg = 0
for i in range(len(vectors)):
for j in range(len(vectors_ul)):
count = count + 1
avg = (count-1)*avg/count + spatial.distance.cosine(vectors[i], vectors_ul[j])/count
avg2 = avg
print('Avg dist b/w ul corner bills, Avg dist b/w ul corner and non-corner bills : ' + str(round(avg1,2)) + ',' + str(round(avg2, 2)))
print('Percentage increase : ' + str(round(((avg2-avg1)/avg1),2)))
count = 0
avg1 = 0.0
avg = 0.0
for i in range(len(vectors_ur)):
for j in range(i+1, len(vectors_ur)):
count = count + 1
avg = (count-1)*avg/count + spatial.distance.cosine(vectors_ur[i], vectors_ur[j])/count
avg1 = avg
count = 0
avg2 = 0
avg = 0
for i in range(len(vectors)):
for j in range(len(vectors_ur)):
count = count + 1
avg = (count-1)*avg/count + spatial.distance.cosine(vectors[i], vectors_ur[j])/count
avg2 = avg
print('Avg dist b/w ur corner bills, Avg dist b/w ur corner and non-corner bills : ' + str(round(avg1,2)) + ',' + str(round(avg2, 2)))
print('Percentage increase : ' + str(round(((avg2-avg1)/avg1),2)))