omlproject/plot.py at master · bargavj/omlproject · GitHub

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
import pickle
import matplotlib.pyplot as plt

results = pickle.load(open('outputPickleData.p', 'rb'))

###### MLP Accuracy w/o sparsity ######
Adam = results["MLP"]["Adam"]["posttrain"][0]['TrainingAccuracies']
SGD = results["MLP"]["SGD"]["posttrain"][0]['TrainingAccuracies']
Adamax = results["MLP"]["Adamax"]["posttrain"][0]['TrainingAccuracies']
RMSprop = results["MLP"]["RMSprop"]["posttrain"][0]['TrainingAccuracies']
plt.plot(range(1,len(Adam)+1), Adam, 'r-+', label='Adam')
plt.plot(range(1,len(Adam)+1), SGD, 'b-', label='SGD')
plt.plot(range(1,len(Adam)+1), Adamax, 'g-.', label='Adamax')
plt.plot(range(1,len(Adam)+1), RMSprop, '-.', label='RMSprop')
plt.xlabel('Epochs')
plt.ylabel('Accuracy')
legend = plt.legend(loc='lower right', shadow=True, fontsize='x-large')
plt.savefig('MLP_acc_nonsparse.png')
plt.close()

###### CNN Accuracy w/o sparsity ######
Adam = results["CNN"]["Adam"]["posttrain"][0]['TrainingAccuracies']
SGD = results["CNN"]["SGD"]["posttrain"][0]['TrainingAccuracies']
Adamax = results["CNN"]["Adamax"]["posttrain"][0]['TrainingAccuracies']
RMSprop = results["CNN"]["RMSprop"]["posttrain"][0]['TrainingAccuracies']
plt.plot(range(1,len(Adam)+1), Adam, 'r-+', label='Adam')
plt.plot(range(1,len(Adam)+1), SGD, 'b-', label='SGD')
plt.plot(range(1,len(Adam)+1), Adamax, 'g-.', label='Adamax')
plt.plot(range(1,len(Adam)+1), RMSprop, '-.', label='RMSprop')
plt.xlabel('Epochs')
plt.ylabel('Accuracy')
legend = plt.legend(loc='lower right', shadow=True, fontsize='x-large')
plt.savefig('CNN_acc_nonsparse.png')
plt.close()

###### MLP Accuracy posttrain sparsity ######
AdamSparsity = [results["MLP"]["Adam"]["posttrain"][i]['sparselevel'] for i in [0, 0.01, 0.025, 0.05, 0.1]]
AdamAccuracy = [results["MLP"]["Adam"]["posttrain"][i]['testaccuracy'] for i in [0, 0.01, 0.025, 0.05, 0.1]]

SGDSparsity = [results["MLP"]["SGD"]["posttrain"][i]['sparselevel'] for i in [0, 0.01, 0.025, 0.05, 0.1]]
SGDAccuracy = [results["MLP"]["SGD"]["posttrain"][i]['testaccuracy'] for i in [0, 0.01, 0.025, 0.05, 0.1]]

AdamaxSparsity = [results["MLP"]["Adamax"]["posttrain"][i]['sparselevel'] for i in [0, 0.01, 0.025, 0.05, 0.1]]
AdamaxAccuracy = [results["MLP"]["Adamax"]["posttrain"][i]['testaccuracy'] for i in [0, 0.01, 0.025, 0.05, 0.1]]

RMSpropSparsity = [results["MLP"]["RMSprop"]["posttrain"][i]['sparselevel'] for i in [0, 0.01, 0.025, 0.05, 0.1]]
RMSpropAccuracy = [results["MLP"]["RMSprop"]["posttrain"][i]['testaccuracy'] for i in [0, 0.01, 0.025, 0.05, 0.1]]

plt.plot(AdamSparsity, AdamAccuracy, 'r-+', label='Adam')
plt.plot(SGDSparsity, SGDAccuracy, 'b-', label='SGD')
plt.plot(AdamaxSparsity, AdamaxAccuracy, 'g-.', label='Adamax')
plt.plot(RMSpropSparsity, RMSpropAccuracy, '-.', label='RMSprop')
plt.xlabel('Sparsity')
plt.ylabel('Accuracy')
legend = plt.legend(loc='lower left', shadow=True, fontsize='x-large')
plt.savefig('MLP_acc_posttrain_sparse.png')
plt.close()

###### CNN Accuracy posttrain sparsity ######
AdamSparsity = [results["CNN"]["Adam"]["posttrain"][i]['sparselevel'] for i in [0, 0.01, 0.025, 0.05, 0.1]]
AdamAccuracy = [results["CNN"]["Adam"]["posttrain"][i]['testaccuracy'] for i in [0, 0.01, 0.025, 0.05, 0.1]]

SGDSparsity = [results["CNN"]["SGD"]["posttrain"][i]['sparselevel'] for i in [0, 0.01, 0.025, 0.05, 0.1]]
SGDAccuracy = [results["CNN"]["SGD"]["posttrain"][i]['testaccuracy'] for i in [0, 0.01, 0.025, 0.05, 0.1]]

AdamaxSparsity = [results["CNN"]["Adamax"]["posttrain"][i]['sparselevel'] for i in [0, 0.01, 0.025, 0.05, 0.1]]
AdamaxAccuracy = [results["CNN"]["Adamax"]["posttrain"][i]['testaccuracy'] for i in [0, 0.01, 0.025, 0.05, 0.1]]

RMSpropSparsity = [results["CNN"]["RMSprop"]["posttrain"][i]['sparselevel'] for i in [0, 0.01, 0.025, 0.05, 0.1]]
RMSpropAccuracy = [results["CNN"]["RMSprop"]["posttrain"][i]['testaccuracy'] for i in [0, 0.01, 0.025, 0.05, 0.1]]

plt.plot(AdamSparsity, AdamAccuracy, 'r-+', label='Adam')
plt.plot(SGDSparsity, SGDAccuracy, 'b-', label='SGD')
plt.plot(AdamaxSparsity, AdamaxAccuracy, 'g-.', label='Adamax')
plt.plot(RMSpropSparsity, RMSpropAccuracy, '-.', label='RMSprop')
plt.xlabel('Sparsity')
plt.ylabel('Accuracy')
legend = plt.legend(loc='lower left', shadow=True, fontsize='x-large')
plt.savefig('CNN_acc_posttrain_sparse.png')
plt.close()

###### MLP Accuracy intrain sparsity ######
AdamSparsity = [results["MLP"]["Adam"]["intrain"][i]['sparselevel'] for i in [0.01, 0.025, 0.05, 0.1]]
AdamSparsity.insert(0, results["MLP"]["Adam"]["posttrain"][0]['sparselevel'])
AdamAccuracy = [results["MLP"]["Adam"]["intrain"][i]['testaccuracy'] for i in [0.01, 0.025, 0.05, 0.1]]
AdamAccuracy.insert(0, results["MLP"]["Adam"]["posttrain"][0]['testaccuracy'])

SGDSparsity = [results["MLP"]["SGD"]["intrain"][i]['sparselevel'] for i in [0.01, 0.025, 0.05, 0.1]]
SGDSparsity.insert(0, results["MLP"]["SGD"]["posttrain"][0]['sparselevel'])
SGDAccuracy = [results["MLP"]["SGD"]["intrain"][i]['testaccuracy'] for i in [0.01, 0.025, 0.05, 0.1]]
SGDAccuracy.insert(0, results["MLP"]["SGD"]["posttrain"][0]['testaccuracy'])

AdamaxSparsity = [results["MLP"]["Adamax"]["intrain"][i]['sparselevel'] for i in [0.01, 0.025, 0.05, 0.1]]
AdamaxSparsity.insert(0, results["MLP"]["Adamax"]["posttrain"][0]['sparselevel'])
AdamaxAccuracy = [results["MLP"]["Adamax"]["intrain"][i]['testaccuracy'] for i in [0.01, 0.025, 0.05, 0.1]]
AdamaxAccuracy.insert(0, results["MLP"]["Adamax"]["posttrain"][0]['testaccuracy'])

RMSpropSparsity = [results["MLP"]["RMSprop"]["intrain"][i]['sparselevel'] for i in [0.01, 0.025, 0.05, 0.1]]
RMSpropSparsity.insert(0, results["MLP"]["RMSprop"]["posttrain"][0]['sparselevel'])
RMSpropAccuracy = [results["MLP"]["RMSprop"]["intrain"][i]['testaccuracy'] for i in [0.01, 0.025, 0.05, 0.1]]
RMSpropAccuracy.insert(0, results["MLP"]["RMSprop"]["posttrain"][0]['testaccuracy'])

plt.plot(AdamSparsity, AdamAccuracy, 'r-+', label='Adam')
plt.plot(SGDSparsity, SGDAccuracy, 'b-', label='SGD')
plt.plot(AdamaxSparsity, AdamaxAccuracy, 'g-.', label='Adamax')
plt.plot(RMSpropSparsity, RMSpropAccuracy, '-.', label='RMSprop')
plt.xlabel('Sparsity')
plt.ylabel('Accuracy')
legend = plt.legend(loc='lower left', shadow=True, fontsize='x-large')
plt.savefig('MLP_acc_intrain_sparse.png')
plt.close()

###### CNN Accuracy intrain sparsity ######
AdamSparsity = [results["CNN"]["Adam"]["intrain"][i]['sparselevel'] for i in [0.01, 0.025, 0.05, 0.1]]
AdamSparsity.insert(0, results["CNN"]["Adam"]["posttrain"][0]['sparselevel'])
AdamAccuracy = [results["CNN"]["Adam"]["intrain"][i]['testaccuracy'] for i in [0.01, 0.025, 0.05, 0.1]]
AdamAccuracy.insert(0, results["CNN"]["Adam"]["posttrain"][0]['testaccuracy'])

SGDSparsity = [results["CNN"]["SGD"]["intrain"][i]['sparselevel'] for i in [0.01, 0.025, 0.05, 0.1]]
SGDSparsity.insert(0, results["CNN"]["SGD"]["posttrain"][0]['sparselevel'])
SGDAccuracy = [results["CNN"]["SGD"]["intrain"][i]['testaccuracy'] for i in [0.01, 0.025, 0.05, 0.1]]
SGDAccuracy.insert(0, results["CNN"]["SGD"]["posttrain"][0]['testaccuracy'])

AdamaxSparsity = [results["CNN"]["Adamax"]["intrain"][i]['sparselevel'] for i in [0.01, 0.025, 0.05, 0.1]]
AdamaxSparsity.insert(0, results["CNN"]["Adamax"]["posttrain"][0]['sparselevel'])
AdamaxAccuracy = [results["CNN"]["Adamax"]["intrain"][i]['testaccuracy'] for i in [0.01, 0.025, 0.05, 0.1]]
AdamaxAccuracy.insert(0, results["CNN"]["Adamax"]["posttrain"][0]['testaccuracy'])

RMSpropSparsity = [results["CNN"]["RMSprop"]["intrain"][i]['sparselevel'] for i in [0.01, 0.025, 0.05, 0.1]]
RMSpropSparsity.insert(0, results["CNN"]["RMSprop"]["posttrain"][0]['sparselevel'])
RMSpropAccuracy = [results["CNN"]["RMSprop"]["intrain"][i]['testaccuracy'] for i in [0.01, 0.025, 0.05, 0.1]]
RMSpropAccuracy.insert(0, results["CNN"]["RMSprop"]["posttrain"][0]['testaccuracy'])

plt.plot(AdamSparsity, AdamAccuracy, 'r-+', label='Adam')
plt.plot(SGDSparsity, SGDAccuracy, 'b-', label='SGD')
plt.plot(AdamaxSparsity, AdamaxAccuracy, 'g-.', label='Adamax')
plt.plot(RMSpropSparsity, RMSpropAccuracy, '-.', label='RMSprop')
plt.xlabel('Sparsity')
plt.ylabel('Accuracy')
legend = plt.legend(loc='lower left', shadow=True, fontsize='x-large')
plt.savefig('CNN_acc_intrain_sparse.png')
plt.close()

###### MLP Accuracy intrain 0.01 sparsity ######
Adam = results["MLP"]["Adam"]["intrain"][0.01]['TrainingAccuracies']
SGD = results["MLP"]["SGD"]["intrain"][0.01]['TrainingAccuracies']
Adamax = results["MLP"]["Adamax"]["intrain"][0.01]['TrainingAccuracies']
RMSprop = results["MLP"]["RMSprop"]["intrain"][0.01]['TrainingAccuracies']
plt.plot(range(1,len(Adam)+1), Adam, 'r-+', label='Adam')
plt.plot(range(1,len(Adam)+1), SGD, 'b-', label='SGD')
plt.plot(range(1,len(Adam)+1), Adamax, 'g-.', label='Adamax')
plt.plot(range(1,len(Adam)+1), RMSprop, '-.', label='RMSprop')
plt.xlabel('Epochs')
plt.ylabel('Accuracy')
legend = plt.legend(loc='lower right', shadow=True, fontsize='x-large')
plt.savefig('MLP_acc_intrain_0.01.png')
plt.close()

###### CNN Accuracy intrain 0.01 sparsity ######
Adam = results["CNN"]["Adam"]["intrain"][0.01]['TrainingAccuracies']
SGD = results["CNN"]["SGD"]["intrain"][0.01]['TrainingAccuracies']
Adamax = results["CNN"]["Adamax"]["intrain"][0.01]['TrainingAccuracies']
RMSprop = results["CNN"]["RMSprop"]["intrain"][0.01]['TrainingAccuracies']
plt.plot(range(1,len(Adam)+1), Adam, 'r-+', label='Adam')
plt.plot(range(1,len(Adam)+1), SGD, 'b-', label='SGD')
plt.plot(range(1,len(Adam)+1), Adamax, 'g-.', label='Adamax')
plt.plot(range(1,len(Adam)+1), RMSprop, '-.', label='RMSprop')
plt.xlabel('Epochs')
plt.ylabel('Accuracy')
legend = plt.legend(loc='lower right', shadow=True, fontsize='x-large')
plt.savefig('CNN_acc_intrain_0.01.png')
plt.close()

###### MLP Accuracy intrain 0.025 sparsity ######
Adam = results["MLP"]["Adam"]["intrain"][0.025]['TrainingAccuracies']
SGD = results["MLP"]["SGD"]["intrain"][0.025]['TrainingAccuracies']
Adamax = results["MLP"]["Adamax"]["intrain"][0.025]['TrainingAccuracies']
RMSprop = results["MLP"]["RMSprop"]["intrain"][0.025]['TrainingAccuracies']
plt.plot(range(1,len(Adam)+1), Adam, 'r-+', label='Adam')
plt.plot(range(1,len(Adam)+1), SGD, 'b-', label='SGD')
plt.plot(range(1,len(Adam)+1), Adamax, 'g-.', label='Adamax')
plt.plot(range(1,len(Adam)+1), RMSprop, '-.', label='RMSprop')
plt.xlabel('Epochs')
plt.ylabel('Accuracy')
legend = plt.legend(loc='lower right', shadow=True, fontsize='x-large')
plt.savefig('MLP_acc_intrain_0.025.png')
plt.close()

###### CNN Accuracy intrain 0.025 sparsity ######
Adam = results["CNN"]["Adam"]["intrain"][0.025]['TrainingAccuracies']
SGD = results["CNN"]["SGD"]["intrain"][0.025]['TrainingAccuracies']
Adamax = results["CNN"]["Adamax"]["intrain"][0.025]['TrainingAccuracies']
RMSprop = results["CNN"]["RMSprop"]["intrain"][0.025]['TrainingAccuracies']
plt.plot(range(1,len(Adam)+1), Adam, 'r-+', label='Adam')
plt.plot(range(1,len(Adam)+1), SGD, 'b-', label='SGD')
plt.plot(range(1,len(Adam)+1), Adamax, 'g-.', label='Adamax')
plt.plot(range(1,len(Adam)+1), RMSprop, '-.', label='RMSprop')
plt.xlabel('Epochs')
plt.ylabel('Accuracy')
legend = plt.legend(loc='lower right', shadow=True, fontsize='x-large')
plt.savefig('CNN_acc_intrain_0.025.png')
plt.close()

###### MLP Accuracy intrain 0.05 sparsity ######
Adam = results["MLP"]["Adam"]["intrain"][0.05]['TrainingAccuracies']
SGD = results["MLP"]["SGD"]["intrain"][0.05]['TrainingAccuracies']
Adamax = results["MLP"]["Adamax"]["intrain"][0.05]['TrainingAccuracies']
RMSprop = results["MLP"]["RMSprop"]["intrain"][0.05]['TrainingAccuracies']
plt.plot(range(1,len(Adam)+1), Adam, 'r-+', label='Adam')
plt.plot(range(1,len(Adam)+1), SGD, 'b-', label='SGD')
plt.plot(range(1,len(Adam)+1), Adamax, 'g-.', label='Adamax')
plt.plot(range(1,len(Adam)+1), RMSprop, '-.', label='RMSprop')
plt.xlabel('Epochs')
plt.ylabel('Accuracy')
legend = plt.legend(loc='lower right', shadow=True, fontsize='x-large')
plt.savefig('MLP_acc_intrain_0.05.png')
plt.close()

###### CNN Accuracy intrain 0.05 sparsity ######
Adam = results["CNN"]["Adam"]["intrain"][0.05]['TrainingAccuracies']
SGD = results["CNN"]["SGD"]["intrain"][0.05]['TrainingAccuracies']
Adamax = results["CNN"]["Adamax"]["intrain"][0.05]['TrainingAccuracies']
RMSprop = results["CNN"]["RMSprop"]["intrain"][0.05]['TrainingAccuracies']
plt.plot(range(1,len(Adam)+1), Adam, 'r-+', label='Adam')
plt.plot(range(1,len(Adam)+1), SGD, 'b-', label='SGD')
plt.plot(range(1,len(Adam)+1), Adamax, 'g-.', label='Adamax')
plt.plot(range(1,len(Adam)+1), RMSprop, '-.', label='RMSprop')
plt.xlabel('Epochs')
plt.ylabel('Accuracy')
legend = plt.legend(loc='lower right', shadow=True, fontsize='x-large')
plt.savefig('CNN_acc_intrain_0.05.png')
plt.close()

###### MLP Accuracy intrain 0.1 sparsity ######
Adam = results["MLP"]["Adam"]["intrain"][0.1]['TrainingAccuracies']
SGD = results["MLP"]["SGD"]["intrain"][0.1]['TrainingAccuracies']
Adamax = results["MLP"]["Adamax"]["intrain"][0.1]['TrainingAccuracies']
RMSprop = results["MLP"]["RMSprop"]["intrain"][0.1]['TrainingAccuracies']
plt.plot(range(1,len(Adam)+1), Adam, 'r-+', label='Adam')
plt.plot(range(1,len(Adam)+1), SGD, 'b-', label='SGD')
plt.plot(range(1,len(Adam)+1), Adamax, 'g-.', label='Adamax')
plt.plot(range(1,len(Adam)+1), RMSprop, '-.', label='RMSprop')
plt.xlabel('Epochs')
plt.ylabel('Accuracy')
legend = plt.legend(loc='lower right', shadow=True, fontsize='x-large')
plt.savefig('MLP_acc_intrain_0.1.png')
plt.close()

###### CNN Accuracy intrain 0.1 sparsity ######
Adam = results["CNN"]["Adam"]["intrain"][0.1]['TrainingAccuracies']
SGD = results["CNN"]["SGD"]["intrain"][0.1]['TrainingAccuracies']
Adamax = results["CNN"]["Adamax"]["intrain"][0.1]['TrainingAccuracies']
RMSprop = results["CNN"]["RMSprop"]["intrain"][0.1]['TrainingAccuracies']
plt.plot(range(1,len(Adam)+1), Adam, 'r-+', label='Adam')
plt.plot(range(1,len(Adam)+1), SGD, 'b-', label='SGD')
plt.plot(range(1,len(Adam)+1), Adamax, 'g-.', label='Adamax')
plt.plot(range(1,len(Adam)+1), RMSprop, '-.', label='RMSprop')
plt.xlabel('Epochs')
plt.ylabel('Accuracy')
legend = plt.legend(loc='lower right', shadow=True, fontsize='x-large')
plt.savefig('CNN_acc_intrain_0.1.png')
plt.close()