In [38]:
from os import environ
environ['optimizer'] = 'Adam'
environ['num_workers']= '2'
environ['batch_size']= str(2048)
environ['n_epochs']= '1200'
environ['batch_norm']= 'True'
environ['loss_func']='MSE'
environ['layers'] = '800 700 600 350 200 180'
environ['dropouts'] = '0.11 '* 6
environ['lr'] = '1e-03'
environ['log'] = 'False'
environ['weight_decay'] = '0.011'
environ['cuda_device'] ='cuda:6'
environ['dataset'] = 'data/speedup_dataset2.pkl'
%run utils.ipynb
In [39]:
train_dl, val_dl, test_dl = train_dev_split(dataset, batch_size, num_workers, log=log)
db = fai.basic_data.DataBunch(train_dl, val_dl, test_dl, device=device)
function329_schedule_13
0
{'computations': {'computations_array': [{'comp_id': 1,
'lhs_data_type': 'p_int32',
'loop_iterators_ids': [2, 3],
'operations_histogram': [[5, 3, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 0],
[0,
0,
0,
0]],
'rhs_accesses': {'accesses': [{'access': [[1,
0,
0],
[0,
1,
1]],
'comp_id': 0},
{'access': [[1,
0,
0],
[0,
1,
-1]],
'comp_id': 0},
{'access': [[1,
0,
1],
[0,
1,
0]],
'comp_id': 0},
{'access': [[1,
0,
1],
[0,
1,
1]],
'comp_id': 0},
{'access': [[1,
0,
1],
[0,
1,
-1]],
'comp_id': 0},
{'access': [[1,
0,
-1],
[0,
1,
0]],
'comp_id': 0},
{'access': [[1,
0,
-1],
[0,
1,
1]],
'comp_id': 0},
{'access': [[1,
0,
-1],
[0,
1,
-1]],
'comp_id': 0},
{'access': [[1,
0,
0],
[0,
1,
0]],
'comp_id': 0}],
'n': 9}}],
'n': 1},
'inputs': {'inputs_array': [{'data_type': 'p_int32',
'input_id': 0,
'loop_iterators_ids': [0, 1]}],
'n': 1},
'iterators': {'iterators_array': [{'it_id': 2,
'lower_bound': 1,
'upper_bound': 1048575},
{'it_id': 3,
'lower_bound': 1,
'upper_bound': 63},
{'it_id': 0,
'lower_bound': 0,
'upper_bound': 1048576},
{'it_id': 1,
'lower_bound': 0,
'upper_bound': 64}],
'n': 4},
'loops': {'loops_array': [{'assignments': {'assignments_array': [], 'n': 0},
'loop_id': 0,
'loop_it': 2,
'parent': -1,
'position': 0},
{'assignments': {'assignments_array': [{'id': 1,
'position': 0}],
'n': 1},
'loop_id': 1,
'loop_it': 3,
'parent': 0,
'position': 0}],
'n': 2},
'seed': 329,
'type': 2}
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
/data/scratch/henni-mohammed/speedup_model/src/data/loop_ast.py in tile(self, loop_id, factor)
260 try:
--> 261 while loop.iterator.id != loop_id:
262 loop = loop.children[0]
AttributeError: 'Computation' object has no attribute 'iterator'
During handling of the above exception, another exception occurred:
NameError Traceback (most recent call last)
<ipython-input-39-ccbb4c277821> in <module>
----> 1 train_dl, val_dl, test_dl = train_dev_split(dataset, batch_size, num_workers, log=log)
2
3 db = fai.basic_data.DataBunch(train_dl, val_dl, test_dl, device=device)
<ipython-input-38-fa1393f8fd16> in train_dev_split(dataset, batch_size, num_workers, log, seed)
108
109 test_size = validation_size = 10000
--> 110 ds = DatasetFromPkl(dataset, maxsize=None, log=log)
111
112 indices = range(len(ds))
/data/scratch/henni-mohammed/speedup_model/src/data/dataset.py in __init__(self, filename, normalized, log, maxsize)
102 program = self.programs[self.program_indexes[i]]
103
--> 104 self.X.append(program.add_schedule(self.schedules[i]).__array__())
105
106
/data/scratch/henni-mohammed/speedup_model/src/data/loop_ast.py in add_schedule(self, schedule)
273 def add_schedule(self, schedule):
274
--> 275 return Loop_AST(self.name, self.dict_repr, schedule)
276
277 def dtype_to_int(self, dtype):
/data/scratch/henni-mohammed/speedup_model/src/data/loop_ast.py in __init__(self, name, dict_repr, schedule)
218
219 if self.schedule:
--> 220 self.apply_schedule()
221
222
/data/scratch/henni-mohammed/speedup_model/src/data/loop_ast.py in apply_schedule(self)
232 if type_ == 'tiling' and binary_schedule[1] == 1:
233 for loop_id, factor in zip(params, factors):
--> 234 self.tile(loop_id, factor)
235
236 elif type_ == 'interchange' and binary_schedule[0] == 1:
/data/scratch/henni-mohammed/speedup_model/src/data/loop_ast.py in tile(self, loop_id, factor)
269 from pprint import pprint
270 pprint(self.dict_repr)
--> 271 exit(1)
272
273 def add_schedule(self, schedule):
NameError: name 'exit' is not defined
In [34]:
input_size = train_dl.dataset.X.shape[1]
output_size = train_dl.dataset.Y.shape[1]
model = None
if batch_norm:
model = Model_BN(input_size, output_size, hidden_sizes=layers_sizes, drops=drops)
else:
model = Model(input_size, output_size)
if loss_func == 'MSE':
criterion = nn.MSELoss()
elif loss_func == 'MAPE':
criterion = mape_criterion
elif loss_func == 'SMAPE':
criterion = smape_criterion
l = fai.Learner(db, model, loss_func=criterion, metrics=[mape_criterion, rmse_criterion])
if optimizer == 'SGD':
l.opt_func = optim.SGD
In [4]:
l = l.load(f"r_speedup_{optimizer}_batch_norm_{batch_norm}_{loss_func}_nlayers_{len(layers_sizes)}_log_{log}")
In [5]:
l.lr_find()
l.recorder.plot()
LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.
In [35]:
l.fit_one_cycle(int(environ['n_epochs']), float(environ['lr']))
Total time: 1:10:08
epoch
train_loss
valid_loss
mape_criterion
rmse_criterion
1
3.119877
4.159297
103.162346
2.038654
2
2.458937
3.360072
151.471848
1.832884
3
2.052949
2.909236
183.127594
1.704980
4
1.763731
2.477645
166.634384
1.572941
5
1.668764
2.382118
159.341629
1.542754
6
1.627714
2.324363
149.311172
1.524337
7
1.576635
2.319100
138.422226
1.521556
8
1.531117
2.168367
143.406693
1.472214
9
1.482670
2.128766
139.532303
1.457496
10
1.464806
2.069212
141.572311
1.438002
11
1.419815
1.908513
145.050201
1.380745
12
1.388305
1.892902
142.724518
1.375559
13
1.372419
1.945301
140.964157
1.394101
14
1.347962
1.836970
141.336197
1.354999
15
1.352326
1.889487
146.588943
1.374369
16
1.309006
1.726306
143.368561
1.313716
17
1.285257
1.807702
139.191101
1.343356
18
1.273544
1.670614
147.955048
1.291968
19
1.252550
1.799533
137.228226
1.341073
20
1.234485
1.657880
142.605225
1.287345
21
1.231580
1.725841
139.279434
1.313315
22
1.210994
1.812722
138.140488
1.346266
23
1.180092
1.626870
132.472198
1.274803
24
1.183105
1.816368
131.703018
1.346385
25
1.168462
1.933239
125.115196
1.389659
26
1.157859
1.678803
131.229202
1.294858
27
1.113932
1.617365
131.479263
1.271243
28
1.097278
1.659424
131.024597
1.288021
29
1.065781
1.623170
129.736969
1.273783
30
1.048911
1.515324
130.019119
1.229141
31
1.075112
1.613534
136.784012
1.269187
32
1.059132
1.761956
128.306671
1.327267
33
1.027710
1.829119
125.620651
1.351971
34
1.026860
1.664926
125.301064
1.290217
35
0.996929
1.740414
124.529526
1.318961
36
0.999108
1.623054
124.079247
1.272657
37
0.964611
1.539746
130.963364
1.240497
38
0.933684
1.429818
122.551216
1.195192
39
0.992827
1.536408
129.641159
1.239398
40
0.981773
1.465253
129.942169
1.209506
41
0.947084
1.402562
127.626076
1.183241
42
0.966133
1.578613
131.963486
1.256302
43
0.952330
1.420017
126.452904
1.190990
44
0.925124
1.693338
124.465675
1.301196
45
0.909009
1.589748
133.088425
1.260656
46
0.871879
1.480783
128.473282
1.216144
47
0.845078
1.808878
125.664925
1.344234
48
0.809716
1.810237
126.078712
1.345164
49
0.813880
1.909651
125.119987
1.381511
50
0.798009
2.087227
124.800697
1.443841
51
0.803258
2.310949
126.656754
1.519828
52
0.793663
1.842795
120.894127
1.356635
53
0.748131
1.998771
126.817635
1.412342
54
0.774604
2.190444
123.942825
1.479187
55
0.779247
1.846167
124.731285
1.358030
56
0.725704
2.059845
123.273560
1.434904
57
0.720434
1.765807
117.120659
1.328202
58
0.720329
2.041100
118.386749
1.427928
59
0.690613
2.042965
120.170677
1.428296
60
0.693025
1.801416
115.312874
1.342025
61
0.714972
1.930613
121.848785
1.388965
62
0.725971
2.227051
117.441597
1.490409
63
0.730101
2.441448
120.224472
1.561794
64
0.714172
1.906287
112.693214
1.379779
65
0.727435
1.441927
117.230087
1.200135
66
0.697677
1.783687
116.422325
1.335412
67
0.691600
2.100202
128.758087
1.448871
68
0.679232
1.931315
116.228691
1.389670
69
0.666700
1.783789
107.531914
1.334751
70
0.671331
1.691642
119.705940
1.300142
71
0.662158
1.939294
114.952324
1.390400
72
0.642722
1.552013
106.849060
1.245666
73
0.647633
1.749076
113.505035
1.322230
74
0.642517
1.775123
120.564323
1.332165
75
0.635705
2.019512
111.315186
1.420244
76
0.637294
1.873256
107.001228
1.368568
77
0.646236
2.115181
107.304977
1.453196
78
0.646668
1.841369
110.191704
1.356517
79
0.620576
1.753481
107.081078
1.323045
80
0.629455
1.880888
106.930351
1.370908
81
0.621386
1.682623
109.700737
1.295767
82
0.623757
2.241790
105.876648
1.496994
83
0.606325
1.865404
113.165451
1.363879
84
0.587264
1.744814
100.405342
1.320108
85
0.625200
1.658385
101.940125
1.287161
86
0.598585
1.828812
119.196465
1.351268
87
0.601319
1.849705
113.748047
1.359325
88
0.611904
1.893726
100.104897
1.375888
89
0.657705
2.001693
107.370010
1.413228
90
0.560772
1.701046
94.244247
1.303616
91
0.563263
1.903952
103.675331
1.379429
92
0.589511
1.639474
107.521523
1.279435
93
0.580688
1.727988
94.814896
1.314033
94
0.561698
1.555581
109.183334
1.246795
95
0.560054
1.541789
98.742035
1.241515
96
0.572685
1.560505
91.998337
1.248443
97
0.550977
1.392285
93.477066
1.179142
98
0.525389
1.276214
96.099472
1.128984
99
0.574497
1.651492
101.072899
1.284708
100
0.520334
1.454352
88.660294
1.205364
101
0.575967
1.275921
94.958481
1.129089
102
0.508792
1.583715
83.560585
1.258188
103
0.492444
1.565348
86.927773
1.250888
104
0.497619
1.439216
88.708839
1.199178
105
0.495516
1.527188
89.432739
1.235717
106
0.473811
1.354015
85.955040
1.162211
107
0.507704
1.518998
91.637001
1.232058
108
0.470779
1.143936
77.236237
1.069528
109
0.447169
1.508342
76.214355
1.227927
110
0.450888
1.303494
76.073448
1.141683
111
0.477067
1.572500
82.641113
1.253288
112
0.451103
1.294602
83.325600
1.137023
113
0.520759
1.368117
89.187782
1.169408
114
0.469473
1.216594
89.349060
1.102633
115
0.553151
1.224327
95.070930
1.106451
116
0.457413
1.175548
78.910126
1.083967
117
0.423293
1.265560
78.267319
1.124370
118
0.437232
1.092538
76.999428
1.044520
119
0.521231
1.223347
86.320915
1.105628
120
0.533385
1.082803
81.785019
1.040331
121
0.498349
0.901337
97.241203
0.949236
122
0.434286
0.986848
91.683472
0.993010
123
0.464103
1.172653
110.596703
1.082308
124
0.440203
0.981711
74.530876
0.990728
125
0.425814
1.198924
78.298599
1.093840
126
0.421047
1.244012
93.545448
1.115153
127
0.416230
1.070482
84.104942
1.034373
128
0.449729
1.112025
111.098602
1.052973
129
0.447012
0.943311
97.194847
0.971178
130
0.418009
0.845380
97.013260
0.919281
131
0.410404
1.010399
84.565224
1.005171
132
0.448645
1.264179
98.708351
1.123976
133
0.413908
0.869661
139.217270
0.932440
134
0.420460
0.947159
105.644623
0.973030
135
0.408223
0.796965
82.158836
0.891908
136
0.394473
0.835953
95.288734
0.913854
137
0.406611
0.849507
94.149872
0.921625
138
0.425264
0.900812
93.289253
0.948786
139
0.400023
0.892658
86.260635
0.944589
140
0.398121
1.028139
91.781929
1.013703
141
0.394824
1.025008
84.343185
1.012194
142
0.399651
0.814258
111.132004
0.902073
143
0.404619
0.890658
87.060867
0.943483
144
0.385045
0.922776
72.100136
0.960007
145
0.429932
0.772217
131.541168
0.878563
146
0.398133
0.819233
84.047897
0.904864
147
0.387536
0.797998
89.299164
0.893178
148
0.405493
0.903610
107.978584
0.949929
149
0.381627
0.816351
99.753426
0.903346
150
0.388079
0.809674
93.143234
0.899640
151
0.391044
0.750890
116.713875
0.865884
152
0.379460
0.770675
77.504097
0.877845
153
0.362160
0.758912
94.316231
0.870944
154
0.375987
0.881909
115.983864
0.939070
155
0.366753
0.805664
98.779236
0.897316
156
0.384802
0.858883
112.863853
0.926681
157
0.377612
0.750217
113.171425
0.866102
158
0.393419
0.806141
83.516937
0.897701
159
0.425923
0.911638
142.913895
0.954731
160
0.402775
0.782084
91.231117
0.884255
161
0.378353
0.792464
73.942009
0.889809
162
0.359988
0.795851
91.402512
0.891850
163
0.406938
0.766372
106.498253
0.874943
164
0.374237
0.759315
74.799362
0.871235
165
0.350853
0.826245
76.598259
0.908713
166
0.361082
0.791901
86.885361
0.889675
167
0.380390
0.906597
111.837212
0.951838
168
0.367490
0.868620
83.970474
0.931216
169
0.422229
0.772067
89.151100
0.878433
170
0.380213
0.835672
77.550385
0.913889
171
0.351024
0.798530
70.564293
0.893419
172
0.337052
0.747141
77.178871
0.863841
173
0.344512
0.839857
84.891747
0.915677
174
0.387699
0.886987
93.223549
0.941500
175
0.379767
0.744459
93.080574
0.862666
176
0.343896
0.781239
72.250282
0.883770
177
0.333400
0.828569
92.265587
0.909945
178
0.353926
0.749404
90.767807
0.865539
179
0.345051
0.765049
91.186295
0.874200
180
0.333451
0.861779
113.888496
0.928206
181
0.350210
0.866608
82.767464
0.930846
182
0.333832
0.788016
77.828529
0.887617
183
0.347594
0.914941
107.235359
0.956304
184
0.345171
0.902789
91.147713
0.950025
185
0.336815
0.891538
106.805000
0.943810
186
0.323646
0.799539
86.381248
0.894110
187
0.324713
0.788986
98.400391
0.888098
188
0.345565
0.826268
102.615425
0.908865
189
0.370211
0.895061
116.022087
0.945962
190
0.331044
0.755006
81.905067
0.868349
191
0.318781
0.758502
61.613487
0.870877
192
0.320465
0.897615
84.835007
0.947216
193
0.316189
0.834078
68.267654
0.912701
194
0.335941
0.764035
90.115723
0.873791
195
0.324193
0.797403
64.965546
0.892728
196
0.359742
1.134844
146.151077
1.065252
197
0.320930
0.860277
80.237473
0.926928
198
0.322555
0.889076
88.259239
0.941705
199
0.304224
0.952582
84.828072
0.975697
200
0.326293
0.780356
90.985100
0.883298
201
0.302418
0.837071
82.402313
0.914552
202
0.320392
0.867054
120.138603
0.930783
203
0.317615
0.929972
69.208946
0.963967
204
0.312235
0.872417
92.632652
0.933760
205
0.296249
0.784462
92.364319
0.885592
206
0.306158
0.816971
94.135246
0.903701
207
0.306155
0.742034
63.927864
0.860868
208
0.301421
0.821611
82.755539
0.906163
209
0.320481
1.103151
122.203796
1.050255
210
0.301008
0.823386
89.938034
0.907202
211
0.300694
1.203958
142.501266
1.097152
212
0.296282
1.029641
122.960411
1.014462
213
0.293435
0.865260
78.518204
0.929680
214
0.286251
0.796625
88.672325
0.892038
215
0.298416
0.942071
94.338516
0.970396
216
0.340669
1.397497
156.087982
1.181653
217
0.318688
1.030097
126.973503
1.014711
218
0.301689
0.853487
71.159897
0.923574
219
0.292013
0.926609
103.604103
0.962503
220
0.288925
0.912864
84.617928
0.954916
221
0.286298
0.847233
79.584534
0.920363
222
0.285012
0.861566
98.047508
0.927763
223
0.368519
2.136118
211.229599
1.460296
224
0.299659
0.811044
76.671349
0.900255
225
0.298598
0.767647
106.283722
0.875768
226
0.293022
0.679274
93.893509
0.823724
227
0.279351
0.816693
80.401688
0.903562
228
0.277205
0.967702
95.106056
0.983426
229
0.277234
1.175310
126.180374
1.082468
230
0.284127
0.784944
88.475197
0.885557
231
0.265250
0.728695
89.943253
0.853067
232
0.301971
0.826045
105.790802
0.908771
233
0.328195
1.273563
149.128143
1.128394
234
0.275396
0.829653
67.621216
0.910761
235
0.283674
0.736321
98.964249
0.857982
236
0.266616
0.737766
84.844200
0.858718
237
0.271107
0.927562
89.297714
0.962715
238
0.273359
0.812995
101.312759
0.901483
239
0.272141
0.886825
92.660278
0.941471
240
0.278152
0.932879
102.626686
0.965766
241
0.270659
0.796869
79.640648
0.892437
242
0.284191
0.773771
98.287148
0.879522
243
0.264793
0.793620
86.187035
0.890427
244
0.257210
0.718339
83.565300
0.847144
245
0.259785
0.764477
71.335388
0.874164
246
0.273157
0.806770
82.735222
0.897866
247
0.253710
0.720641
76.467751
0.848815
248
0.262025
0.954865
96.325684
0.976859
249
0.268407
0.884943
76.048065
0.940424
250
0.298947
0.916009
133.181168
0.956657
251
0.318131
0.913326
108.592072
0.955611
252
0.296110
1.260913
167.714600
1.122329
253
0.276979
0.620239
113.235184
0.787310
254
0.277610
0.660697
86.102692
0.812336
255
0.291700
1.265321
134.239273
1.124314
256
0.280236
0.653201
86.957199
0.807950
257
0.272288
0.763280
114.304062
0.873640
258
0.249816
0.660259
75.924301
0.812333
259
0.290930
1.174361
148.479538
1.083418
260
0.264436
0.746237
85.929703
0.863457
261
0.247780
0.741691
105.091949
0.860598
262
0.260235
0.693707
101.073166
0.832639
263
0.253443
0.763241
90.903900
0.873245
264
0.259091
0.807865
96.762428
0.898432
265
0.248276
0.838976
85.648918
0.915445
266
0.251074
0.751494
96.106934
0.866626
267
0.250623
0.757178
113.887909
0.869895
268
0.266792
0.725149
95.338615
0.851479
269
0.235772
0.632280
67.180054
0.795018
270
0.241555
0.645678
80.581436
0.803055
271
0.243842
0.855396
116.358177
0.924776
272
0.253437
0.780699
93.213730
0.883486
273
0.246309
0.756766
84.243500
0.869872
274
0.256385
0.706212
94.465553
0.840257
275
0.234727
0.919730
79.999413
0.958781
276
0.269743
0.869879
125.565125
0.932450
277
0.250610
0.812189
105.235085
0.900941
278
0.253645
0.802904
78.564682
0.895995
279
0.249776
0.771783
77.184891
0.878124
280
0.283183
1.627154
187.646851
1.275392
281
0.262117
0.773728
87.275063
0.879138
282
0.307278
1.444251
144.313629
1.201330
283
0.280550
0.867839
105.944984
0.930639
284
0.251906
0.644228
93.046761
0.802452
285
0.266842
1.499736
120.020950
1.224316
286
0.248090
0.703440
76.797653
0.837846
287
0.245094
0.727104
102.375877
0.852635
288
0.230488
0.736103
70.846344
0.857377
289
0.234458
0.716166
83.603340
0.845856
290
0.246291
0.764694
84.420151
0.874256
291
0.230799
0.669530
82.840904
0.818094
292
0.222857
0.655805
102.233360
0.809741
293
0.234974
0.802511
78.100540
0.895122
294
0.229592
0.634351
71.844833
0.796203
295
0.228255
0.634107
63.998569
0.795928
296
0.223907
0.688829
98.157127
0.829839
297
0.231711
0.575030
78.714806
0.757879
298
0.245472
0.881846
70.973351
0.939009
299
0.228500
0.788420
69.483078
0.887438
300
0.226099
0.652906
67.422112
0.807895
301
0.233976
0.731523
87.110451
0.854743
302
0.227537
0.796078
85.625008
0.892038
303
0.228869
0.699666
68.950256
0.836060
304
0.235716
0.868270
128.320648
0.931443
305
0.229153
0.899539
124.649498
0.948303
306
0.228675
0.833931
85.563377
0.911846
307
0.225483
0.905352
98.984665
0.951017
308
0.242498
0.697858
82.813477
0.835166
309
0.233908
0.846067
98.336853
0.919692
310
0.233159
0.784875
105.450439
0.885717
311
0.217862
0.628170
65.930382
0.792173
312
0.222727
0.678630
78.981316
0.823424
313
0.216725
0.808236
106.970474
0.898971
314
0.224687
0.745103
81.101463
0.862802
315
0.235847
0.768987
75.749054
0.876712
316
0.226577
15.622602
211.217300
3.929859
317
0.232373
0.683162
76.555077
0.826323
318
0.221976
0.774872
93.447609
0.880079
319
0.225281
1.011397
93.219055
1.005584
320
0.245275
0.826487
88.257790
0.908305
321
0.224615
1.007878
91.460358
1.003521
322
0.231811
0.900651
109.444824
0.948605
323
0.223130
0.847339
77.097298
0.920303
324
0.219096
0.949094
87.974739
0.974052
325
0.221336
0.762967
74.311172
0.872458
326
0.220856
0.861768
95.128334
0.928116
327
0.244272
0.911336
68.885872
0.952458
328
0.262224
1.250535
130.411423
1.117355
329
0.235260
0.770673
99.912712
0.877648
330
0.228126
0.956489
109.712799
0.977811
331
0.232188
1.319135
136.030334
1.147599
332
0.247319
0.645090
76.824562
0.802792
333
0.244676
1.998824
163.269867
1.413252
334
0.228852
0.816678
87.424797
0.903571
335
0.238165
0.777065
81.270363
0.881305
336
0.240483
0.933654
110.913475
0.965763
337
0.226187
0.873136
105.647102
0.933365
338
0.234225
0.896849
110.653839
0.946418
339
0.227547
0.902291
98.798752
0.949272
340
0.228916
0.831217
83.480820
0.911668
341
0.222605
1.130754
123.523712
1.063292
342
0.219920
0.834561
116.519211
0.913157
343
0.225831
0.666759
80.182159
0.816261
344
0.218172
0.681363
95.342041
0.825384
345
0.220457
0.657926
70.505379
0.810817
346
0.230048
1.177985
130.895172
1.084875
347
0.223617
0.684028
92.021797
0.826309
348
0.222605
0.694678
97.994644
0.833326
349
0.210572
0.690080
73.610909
0.830484
350
0.218730
0.689496
99.281265
0.830043
351
0.213774
0.691322
69.995178
0.830887
352
0.210645
0.683251
84.802574
0.825961
353
0.210441
0.764333
88.818672
0.874056
354
0.214490
0.941601
79.872185
0.969998
355
0.213130
0.639263
69.442726
0.799124
356
0.216927
0.983361
115.218529
0.990951
357
0.202668
0.647769
64.058327
0.804598
358
0.203280
1.040189
119.565422
1.019660
359
0.212284
0.660704
74.146828
0.812656
360
0.211444
0.928285
112.015274
0.963301
361
0.206098
0.683867
72.134850
0.826672
362
0.216711
0.783290
82.926521
0.884910
363
0.203371
0.606602
58.611320
0.778778
364
0.202435
0.791825
104.606224
0.889602
365
0.213709
0.615473
66.120506
0.784027
366
0.213674
0.758017
76.182594
0.870440
367
0.199344
0.675579
69.926003
0.821750
368
0.192811
0.841059
70.954391
0.916882
369
0.208975
1.145176
117.890213
1.069797
370
0.200861
0.719430
86.825233
0.847988
371
0.195095
0.667870
64.505699
0.817129
372
0.206399
0.775960
75.239235
0.880646
373
0.218792
1.019172
124.695427
1.009088
374
0.204049
0.806153
96.917732
0.897609
375
0.199696
0.756927
95.658836
0.869496
376
0.200408
0.768328
88.711365
0.876287
377
0.200879
0.828363
74.565559
0.910024
378
0.208887
0.844610
85.011841
0.918878
379
0.214161
1.626636
107.404961
1.274368
380
0.200820
0.768562
78.414589
0.876293
381
0.199092
0.668663
57.640812
0.817478
382
0.203397
0.949211
106.370773
0.974119
383
0.197458
0.797059
71.298500
0.892024
384
0.198394
0.730811
81.077950
0.854727
385
0.193977
0.734650
78.361176
0.856895
386
0.196510
0.828459
102.868523
0.909892
387
0.199615
0.838440
78.512962
0.914782
388
0.197338
0.837759
74.124123
0.915020
389
0.206669
0.671283
76.024216
0.819102
390
0.195138
0.791319
80.677017
0.889095
391
0.194463
0.701071
69.961319
0.837283
392
0.195832
0.782509
85.922096
0.884590
393
0.191890
0.960386
98.123322
0.979700
394
0.191554
0.725894
68.308273
0.851883
395
0.190876
0.778486
65.531944
0.881895
396
0.194399
0.671020
68.840775
0.818664
397
0.208835
0.982793
95.806450
0.991307
398
0.205751
0.705977
56.438782
0.840087
399
0.195821
0.784622
78.864128
0.885566
400
0.188593
0.694300
61.621632
0.833049
401
0.190864
0.765340
86.861610
0.874365
402
0.197583
0.705862
70.316513
0.839699
403
0.192719
0.680899
72.477875
0.825072
404
0.185092
1.104765
82.147903
1.050566
405
0.199031
0.878106
93.645966
0.936838
406
0.189043
0.816121
79.173332
0.902903
407
0.191951
0.802793
72.359047
0.895661
408
0.194524
0.816260
92.267502
0.903402
409
0.217667
1.394414
115.777298
1.180494
410
0.202061
1.276918
116.471886
1.129597
411
0.200025
0.803632
89.230888
0.896193
412
0.210458
1.386700
143.310989
1.177353
413
0.199641
1.200805
108.483002
1.095563
414
0.193242
0.682016
73.321884
0.825640
415
0.199728
1.624444
116.561096
1.274221
416
0.209415
0.902961
94.905296
0.949960
417
0.198016
0.954730
91.028816
0.976700
418
0.194569
1.275772
95.473236
1.128737
419
0.192882
0.842065
79.687355
0.916535
420
0.192463
1.276895
101.503540
1.129912
421
0.195352
0.857146
79.774940
0.925488
422
0.224396
1.492076
121.577148
1.221066
423
0.200649
0.977463
89.666359
0.988020
424
0.199738
1.514442
113.748985
1.230394
425
0.204107
1.118153
107.328751
1.057078
426
0.195197
0.740241
68.909721
0.859928
427
0.200876
1.407341
158.468643
1.186089
428
0.197815
1.395509
87.630188
1.180571
429
0.203609
0.785674
86.542999
0.885877
430
0.200887
1.272331
91.055969
1.127928
431
0.193998
0.963452
77.150963
0.980694
432
0.189163
0.969940
84.530708
0.984667
433
0.190898
1.807559
134.333206
1.342594
434
0.194543
1.338659
111.516174
1.156371
435
0.197728
0.953306
89.595848
0.976287
436
0.194299
1.428059
98.974152
1.194585
437
0.197614
1.083766
84.498512
1.040805
438
0.188209
1.430697
107.003036
1.195971
439
0.184805
0.867055
75.694725
0.930839
440
0.195644
1.163394
84.388885
1.078101
441
0.194371
1.686524
114.293236
1.297789
442
0.188787
1.573894
120.879585
1.253795
443
0.188109
1.107702
86.374893
1.052023
444
0.184588
1.470064
107.167923
1.212154
445
0.200230
1.630721
118.229691
1.276893
446
0.197946
1.040828
90.391953
1.019076
447
0.191403
1.458320
101.082573
1.207271
448
0.192630
1.130447
85.335304
1.062794
449
0.185521
0.998909
98.818504
0.999294
450
0.181756
0.915229
92.091148
0.956397
451
0.185808
0.909141
73.085854
0.952944
452
0.181819
0.714018
70.079498
0.844202
453
0.191848
1.529204
111.991264
1.235726
454
0.179765
0.729545
75.495102
0.853801
455
0.188978
1.748733
102.679642
1.321902
456
0.185963
1.268744
86.292648
1.125784
457
0.187214
1.352678
107.565971
1.161716
458
0.184912
1.347936
86.267044
1.160743
459
0.181287
1.348074
100.179665
1.160834
460
0.184653
1.312083
97.856133
1.144514
461
0.182777
1.199746
95.485046
1.094271
462
0.196325
2.440196
147.396469
1.561132
463
0.190025
1.292557
99.104477
1.135500
464
0.181765
1.608309
99.724480
1.267494
465
0.182783
1.230560
112.480904
1.109152
466
0.184785
1.616660
117.705177
1.270670
467
0.184853
0.897237
83.536575
0.947084
468
0.189486
1.945063
129.685318
1.394501
469
0.177805
1.565519
90.409340
1.251149
470
0.182446
1.238222
80.975838
1.112377
471
0.180425
1.063235
79.952621
1.030718
472
0.180596
0.968374
92.617935
0.983691
473
0.176510
1.846807
105.588074
1.358179
474
0.180593
1.381484
95.267525
1.175079
475
0.180575
2.565567
136.303482
1.600059
476
0.185613
1.132393
103.884956
1.063522
477
0.184088
0.989386
89.559685
0.994401
478
0.190501
1.138399
104.365723
1.066695
479
0.181047
1.452733
95.982689
1.204971
480
0.176096
0.854483
74.355247
0.924183
481
0.187097
1.156760
88.985497
1.075455
482
0.178684
0.930799
81.112732
0.964226
483
0.178025
1.279528
90.406868
1.131094
484
0.186391
1.364727
86.424973
1.167230
485
0.186354
0.983095
79.160904
0.991163
486
0.177544
1.009420
90.745972
1.003844
487
0.184889
1.814551
125.995125
1.346287
488
0.179484
0.764305
78.602089
0.873705
489
0.186055
1.748083
122.892578
1.321756
490
0.175396
1.074527
96.704437
1.036475
491
0.175129
0.938322
88.498825
0.968572
492
0.175310
1.354279
92.387299
1.162028
493
0.185549
1.054936
106.970879
1.026830
494
0.182745
1.450891
83.892860
1.204098
495
0.178963
1.367628
90.692406
1.169233
496
0.173149
1.356334
103.236328
1.164285
497
0.173450
1.060094
82.456886
1.029514
498
0.174219
1.459061
100.449722
1.207687
499
0.173704
1.157011
75.191353
1.074435
500
0.174038
1.495481
119.946777
1.222289
501
0.186892
1.546335
92.522652
1.242902
502
0.180408
0.992134
73.817253
0.995926
503
0.186944
1.204891
94.750137
1.097338
504
0.177899
0.879116
82.517441
0.937342
505
0.174321
1.393220
99.108917
1.180024
506
0.180834
3.222656
175.369171
1.794385
507
0.181181
0.709289
74.663620
0.841971
508
0.177436
0.728459
78.443382
0.853275
509
0.182386
1.698713
119.007912
1.302674
510
0.172020
0.982239
97.713974
0.990470
511
0.174497
1.064309
101.095932
1.031364
512
0.175292
1.445302
128.433716
1.201857
513
0.180888
1.423756
108.644096
1.192465
514
0.180158
1.343325
103.350304
1.158688
515
0.186335
1.338025
121.494598
1.156510
516
0.175300
0.819619
79.574135
0.905202
517
0.177099
1.356266
117.208115
1.163777
518
0.183087
0.958503
102.014236
0.978628
519
0.181763
0.772579
77.895515
0.878347
520
0.177078
1.123198
101.700752
1.059464
521
0.186109
1.422620
95.302391
1.192315
522
0.175709
1.085071
76.309303
1.041582
523
0.173821
1.046983
92.093567
1.022593
524
0.171753
1.341048
95.993599
1.157028
525
0.178343
0.963177
89.293823
0.981205
526
0.173930
1.298919
83.055031
1.138601
527
0.169903
1.398625
92.041092
1.182487
528
0.177049
1.458302
89.575829
1.207333
529
0.174228
0.827934
73.972015
0.909607
530
0.184908
1.851847
112.942062
1.360368
531
0.182411
1.187760
84.653076
1.089554
532
0.175818
1.081256
84.001236
1.039407
533
0.175517
1.181890
97.405594
1.086920
534
0.169081
0.891171
85.892960
0.943273
535
0.171386
0.863105
87.835976
0.928972
536
0.175707
1.639969
110.209938
1.279721
537
0.173615
0.985973
91.514740
0.992791
538
0.169196
0.692376
77.770988
0.831688
539
0.163958
1.248190
110.881721
1.117083
540
0.167067
1.100993
111.580688
1.049179
541
0.175126
1.381462
99.215752
1.174238
542
0.172896
1.115579
86.352173
1.055823
543
0.167458
0.853393
85.671524
0.923698
544
0.162702
0.788280
76.614578
0.887406
545
0.170040
1.239248
111.969002
1.113158
546
0.173360
0.881492
79.143166
0.938393
547
0.174084
1.245871
116.359512
1.116107
548
0.167671
0.804089
70.485222
0.896685
549
0.161612
0.945899
85.004272
0.971976
550
0.170401
1.189675
102.994141
1.090028
551
0.164247
0.787654
80.033043
0.887181
552
0.166723
1.239319
118.600227
1.113147
553
0.162995
0.980513
81.865952
0.989874
554
0.161717
1.015110
84.519547
1.007490
555
0.162557
0.645522
66.945801
0.803352
556
0.166643
0.777428
80.095894
0.881206
557
0.164953
0.828415
80.857529
0.909772
558
0.173488
0.760314
65.886589
0.871913
559
0.165651
0.928753
84.067307
0.963160
560
0.161421
0.871313
75.573616
0.932902
561
0.165039
1.097705
103.571365
1.047165
562
0.169008
0.957982
77.245499
0.978638
563
0.169006
0.910113
80.223503
0.953827
564
0.176940
1.085906
103.012650
1.041720
565
0.174813
1.568586
132.739014
1.251463
566
0.164756
0.920890
90.711456
0.959602
567
0.160809
0.762684
83.335403
0.872966
568
0.165084
1.107860
98.265869
1.052227
569
0.159698
0.876365
97.461403
0.936102
570
0.165219
0.849489
87.569984
0.921202
571
0.169177
0.902228
95.469139
0.948554
572
0.167667
0.888402
87.948723
0.941447
573
0.165650
0.675177
71.734947
0.821290
574
0.165729
1.149327
102.096809
1.071619
575
0.159606
0.886336
92.920746
0.941272
576
0.167019
0.928675
82.104568
0.963229
577
0.169842
0.851142
76.298653
0.922181
578
0.170249
1.209365
119.416328
1.099570
579
0.165767
0.881788
83.991089
0.938862
580
0.164429
0.884588
90.144722
0.940431
581
0.170302
0.820458
76.562866
0.905211
582
0.169314
1.344414
110.785896
1.158435
583
0.159306
0.615489
62.146336
0.784337
584
0.158697
0.694825
64.825104
0.833131
585
0.159020
0.736898
78.136421
0.858325
586
0.163488
0.663598
63.973255
0.814080
587
0.161267
0.637296
56.016014
0.797731
588
0.157175
0.771841
68.417297
0.878366
589
0.163511
0.813992
61.268864
0.902092
590
0.164548
0.875572
73.753685
0.935618
591
0.166559
1.298104
107.278816
1.137848
592
0.166731
0.818294
90.592865
0.904463
593
0.156793
0.663728
65.990860
0.814412
594
0.161818
0.919696
115.022324
0.958819
595
0.158982
0.826449
80.383240
0.908598
596
0.165855
1.781349
121.685402
1.334434
597
0.160638
0.625215
56.052483
0.790191
598
0.179631
1.284225
94.172989
1.132477
599
0.161516
1.021843
108.914314
1.010665
600
0.159607
0.989508
93.361916
0.994467
601
0.158173
0.909078
81.680870
0.953261
602
0.156304
1.417018
132.042892
1.189611
603
0.158901
1.104166
102.469040
1.050051
604
0.159611
1.159341
102.258743
1.076596
605
0.156230
0.890880
81.038513
0.943516
606
0.159361
1.151249
100.354088
1.072655
607
0.163436
0.858488
65.402428
0.926456
608
0.155602
0.800095
72.227028
0.893997
609
0.161896
0.802588
79.137459
0.895788
610
0.160477
0.927909
82.901131
0.962917
611
0.155754
0.880987
85.682076
0.938291
612
0.154068
0.889208
90.291481
0.942148
613
0.155726
0.928151
76.374710
0.962702
614
0.170776
1.111658
91.013016
1.053956
615
0.163195
0.949262
84.609047
0.974272
616
0.161378
1.235794
100.344872
1.110694
617
0.160325
0.670692
60.027176
0.818550
618
0.162426
0.968751
83.175400
0.983696
619
0.157808
0.693224
62.780613
0.832578
620
0.160107
1.035206
104.129547
1.016868
621
0.158575
0.868503
82.831894
0.931735
622
0.152495
1.209234
93.036079
1.099016
623
0.157412
1.276269
95.820847
1.129299
624
0.161852
0.598958
64.958000
0.773299
625
0.155155
0.752744
74.114578
0.867288
626
0.159236
0.718920
60.157246
0.847709
627
0.149817
0.690380
72.047791
0.830858
628
0.157312
0.939514
75.384186
0.969003
629
0.154122
0.693624
60.914936
0.832684
630
0.154418
0.778047
78.333038
0.882039
631
0.159416
0.784638
69.724823
0.885626
632
0.152666
0.865819
74.132729
0.929176
633
0.158930
1.517493
110.503998
1.231149
634
0.156463
1.001414
79.592262
1.000091
635
0.153745
0.757459
69.522987
0.869902
636
0.172636
1.394411
134.424530
1.180526
637
0.155642
1.083498
87.976082
1.040447
638
0.148952
0.772564
74.042213
0.877905
639
0.153621
0.724997
72.397278
0.850996
640
0.156874
0.857371
77.512489
0.925332
641
0.148590
0.831811
80.805290
0.911329
642
0.160310
1.129749
123.900185
1.062321
643
0.151881
0.875648
89.534515
0.935546
644
0.151984
0.762671
72.876434
0.873206
645
0.151030
0.749866
82.375885
0.864668
646
0.149223
0.686414
59.858189
0.828228
647
0.150938
0.709556
73.134125
0.841987
648
0.155738
0.826865
80.177406
0.908886
649
0.150255
0.748767
71.170937
0.864898
650
0.150702
0.890812
92.918098
0.943538
651
0.150025
0.891806
83.293816
0.943636
652
0.153230
0.919228
86.778717
0.958124
653
0.149904
0.790074
75.985062
0.888804
654
0.148300
0.975401
90.183228
0.987492
655
0.150710
0.892328
86.170319
0.944050
656
0.147735
1.062493
85.903252
1.030249
657
0.163327
1.079081
106.893814
1.038650
658
0.150316
0.849465
84.126328
0.921221
659
0.149487
0.789612
82.865158
0.888204
660
0.145025
0.803735
68.724777
0.896395
661
0.147066
0.824219
80.654984
0.907244
662
0.145528
0.850705
83.568024
0.921799
663
0.150325
0.844134
84.048813
0.918041
664
0.152695
0.753228
70.684982
0.867691
665
0.156302
0.945353
93.688828
0.971265
666
0.152846
1.016449
84.211159
1.007983
667
0.149545
0.829023
70.909149
0.910293
668
0.148552
0.790032
69.049629
0.888538
669
0.153924
0.869687
79.376213
0.932023
670
0.149585
0.813574
69.518509
0.900836
671
0.153031
0.942514
90.482330
0.970443
672
0.149872
0.787789
73.426529
0.887213
673
0.145407
0.830929
73.761185
0.911256
674
0.150055
0.875971
84.523293
0.935666
675
0.148884
1.132704
84.826317
1.063916
676
0.153802
0.704650
62.244461
0.838914
677
0.145866
0.857977
79.679253
0.925616
678
0.148143
0.844393
78.162659
0.918450
679
0.148465
1.039542
90.849503
1.019187
680
0.146271
0.732675
70.579155
0.855720
681
0.145160
0.949075
99.950027
0.973675
682
0.147077
0.831624
74.555153
0.911648
683
0.151375
0.643900
66.422646
0.802242
684
0.151483
0.796277
74.360428
0.891806
685
0.146627
0.698182
68.162216
0.835113
686
0.149764
0.716602
74.722153
0.846164
687
0.151648
0.683345
62.773636
0.826510
688
0.147346
0.829545
81.886635
0.910592
689
0.146854
0.741642
75.199654
0.860839
690
0.148829
0.827010
83.582703
0.909141
691
0.148929
0.759049
73.489609
0.870719
692
0.149247
0.858050
94.313614
0.926094
693
0.143388
0.829318
82.708641
0.910499
694
0.146423
0.712063
58.106533
0.843718
695
0.143257
0.792316
71.407303
0.889872
696
0.139467
0.765836
73.910751
0.874778
697
0.146959
0.900611
96.505646
0.948869
698
0.149362
0.987489
87.820984
0.993545
699
0.141060
0.757943
72.608871
0.870046
700
0.139828
0.821429
80.135277
0.905901
701
0.140958
0.784438
81.769402
0.885268
702
0.142905
0.741616
84.047684
0.860914
703
0.142083
0.730264
65.630562
0.854375
704
0.140067
1.159846
96.942413
1.076026
705
0.145036
0.658421
70.390045
0.811147
706
0.141408
0.762116
77.951202
0.872885
707
0.145069
0.763490
71.338936
0.872823
708
0.150596
0.797016
76.652786
0.892649
709
0.142663
0.652987
70.484062
0.807737
710
0.143505
0.959911
96.830772
0.979024
711
0.140927
0.913673
89.596581
0.955090
712
0.143370
0.953766
94.438210
0.976516
713
0.146322
0.698682
73.057991
0.834927
714
0.142214
0.853727
78.537308
0.923277
715
0.140925
0.849873
85.139145
0.921650
716
0.151273
0.933031
90.031853
0.965529
717
0.148344
0.921854
88.754997
0.960020
718
0.148993
0.964579
90.154663
0.981587
719
0.149809
1.098740
91.298653
1.047737
720
0.145542
0.979445
88.924911
0.988291
721
0.141002
0.700170
69.151314
0.836650
722
0.145131
0.867658
85.411247
0.931320
723
0.146706
0.770292
65.098785
0.877275
724
0.145645
0.694163
65.057816
0.832691
725
0.147690
0.766687
78.103035
0.875549
726
0.143415
0.795148
74.662567
0.891419
727
0.143554
0.846380
87.079872
0.918749
728
0.146870
0.894122
86.347122
0.944690
729
0.147715
0.788871
75.539009
0.887916
730
0.147313
0.930390
83.511955
0.964119
731
0.140192
0.868210
82.070747
0.931601
732
0.142652
0.862561
80.056557
0.928380
733
0.144782
0.859035
80.776138
0.926671
734
0.141168
0.738665
70.779442
0.859357
735
0.139451
0.668353
68.009804
0.817030
736
0.134980
0.966177
92.350700
0.982547
737
0.138741
0.830810
70.386642
0.911087
738
0.138229
0.753456
70.733963
0.867528
739
0.138013
0.833549
77.222694
0.912951
740
0.138158
0.732452
74.050972
0.855684
741
0.143803
0.673194
55.675224
0.820239
742
0.140522
0.788183
70.766106
0.887612
743
0.143405
0.691782
70.987053
0.831460
744
0.142258
0.895171
85.181229
0.945632
745
0.139458
0.717190
72.829903
0.846737
746
0.135828
0.772244
63.699245
0.878534
747
0.141145
0.736908
60.455173
0.857904
748
0.145057
0.909786
86.077454
0.953401
749
0.141895
0.730835
55.053806
0.854584
750
0.136927
0.842168
83.019073
0.917648
751
0.140459
0.912307
80.909286
0.954241
752
0.140977
0.838125
80.370262
0.915179
753
0.138222
0.855136
92.377686
0.924563
754
0.141385
0.938021
92.143814
0.968454
755
0.139170
0.637330
49.196426
0.798256
756
0.135967
1.070291
80.122086
1.033933
757
0.141540
0.629437
56.767380
0.792818
758
0.139181
0.610211
60.455395
0.780746
759
0.134261
0.626026
60.625820
0.791090
760
0.145350
0.966435
84.377327
0.982720
761
0.141219
0.782872
69.419189
0.884536
762
0.141002
0.722433
73.031761
0.849607
763
0.133006
0.720296
68.477715
0.848302
764
0.134138
0.786528
67.884903
0.886506
765
0.136377
0.805257
62.704475
0.896570
766
0.135813
0.679017
55.923164
0.823736
767
0.137290
0.874572
85.642273
0.934445
768
0.139021
0.722816
52.543537
0.849280
769
0.134938
0.721796
59.651924
0.849145
770
0.137956
0.672354
55.297367
0.819780
771
0.133038
0.672657
60.190861
0.820014
772
0.132928
0.707819
62.137955
0.840831
773
0.135470
0.760348
64.833504
0.871517
774
0.136777
0.782569
75.058495
0.884159
775
0.134267
0.595628
66.217697
0.771453
776
0.137450
0.810084
81.734398
0.900009
777
0.139154
0.834705
75.696297
0.913357
778
0.137214
0.803814
74.287003
0.896455
779
0.136605
0.855918
76.202576
0.925074
780
0.132328
0.793015
73.810226
0.889253
781
0.136068
0.725413
63.327805
0.851334
782
0.134998
0.735236
79.457085
0.857234
783
0.137037
0.815389
68.237335
0.902677
784
0.141699
0.735451
61.698601
0.857267
785
0.143027
0.883287
84.608551
0.939326
786
0.138913
0.715244
66.618179
0.845338
787
0.137357
0.743260
68.992958
0.862037
788
0.134902
0.782906
77.621887
0.884476
789
0.132085
0.755246
67.271080
0.868662
790
0.136043
0.690477
71.485229
0.830748
791
0.130810
0.642984
56.834675
0.801487
792
0.133258
0.780046
75.070984
0.882963
793
0.134483
0.739007
68.777054
0.859553
794
0.132461
0.727079
69.103584
0.852639
795
0.132995
0.781789
70.328239
0.884096
796
0.137530
0.697417
56.050949
0.834785
797
0.138565
0.918546
90.393265
0.958260
798
0.142959
0.803625
65.658516
0.896293
799
0.134219
0.827149
82.063705
0.908917
800
0.135272
0.807847
82.009331
0.898407
801
0.137839
0.759706
75.174141
0.870845
802
0.133868
0.681547
62.379974
0.825324
803
0.133795
0.824020
84.770927
0.907205
804
0.132306
0.829868
71.151321
0.910714
805
0.132891
0.739162
70.403473
0.859366
806
0.140140
0.804481
82.427689
0.896211
807
0.135848
0.812914
74.971100
0.901411
808
0.136145
0.770378
76.852348
0.877685
809
0.132427
0.986331
105.018288
0.992893
810
0.131354
0.812691
81.480278
0.901344
811
0.129834
0.832237
71.475166
0.911851
812
0.131838
0.783876
77.681763
0.884866
813
0.131571
0.753372
74.308540
0.867688
814
0.133947
0.711661
72.192642
0.843313
815
0.132863
0.809984
80.340225
0.899666
816
0.127983
0.832700
80.468849
0.912282
817
0.135596
0.817659
67.471581
0.903779
818
0.127992
0.800468
72.416512
0.894554
819
0.135223
0.799768
75.606560
0.894012
820
0.130051
0.788218
70.451294
0.886550
821
0.132455
0.859685
82.569725
0.925971
822
0.128277
0.930151
88.491119
0.964150
823
0.130052
0.738669
69.206551
0.858968
824
0.127728
0.685790
70.349274
0.827884
825
0.127141
0.973838
82.392159
0.986199
826
0.126370
0.671615
78.842697
0.819178
827
0.129487
0.712044
64.748306
0.843302
828
0.125942
0.681542
63.548492
0.824952
829
0.132295
0.710244
73.662628
0.842581
830
0.132181
0.685735
61.087967
0.827791
831
0.129376
0.732298
70.953171
0.855650
832
0.131665
0.655070
53.336552
0.809148
833
0.129135
0.767638
71.542015
0.875895
834
0.127851
0.738866
69.451195
0.859400
835
0.128787
0.763613
81.898903
0.873193
836
0.135071
0.874869
81.566391
0.935138
837
0.133684
0.827637
74.299240
0.908775
838
0.129206
0.769226
70.697342
0.876766
839
0.129678
0.641792
54.961533
0.800852
840
0.124755
0.679752
60.352402
0.824387
841
0.131354
0.754859
63.659889
0.868705
842
0.130730
0.803999
67.010056
0.895996
843
0.129934
0.765631
73.773239
0.874789
844
0.129745
0.726068
64.114990
0.851836
845
0.131370
0.693417
64.278412
0.832478
846
0.129534
0.791103
70.968597
0.889275
847
0.126288
0.817280
80.383675
0.903434
848
0.125069
0.855840
87.548981
0.924328
849
0.129508
0.802894
75.846062
0.895558
850
0.127506
0.806673
78.925148
0.897788
851
0.126308
0.770416
70.535324
0.877619
852
0.124011
0.785654
73.624588
0.886212
853
0.122685
0.860758
80.499603
0.927246
854
0.130077
0.934538
86.273361
0.965886
855
0.130061
0.787638
73.446999
0.887382
856
0.127646
0.792422
79.226295
0.889912
857
0.125300
0.699591
69.284447
0.835967
858
0.125912
0.820846
68.813934
0.905570
859
0.126598
0.686241
53.014824
0.827954
860
0.128294
0.848718
73.271309
0.921079
861
0.129246
0.760581
62.571987
0.871836
862
0.127758
0.797367
79.280708
0.892465
863
0.125410
0.840746
70.574196
0.916674
864
0.124489
0.779022
69.533325
0.882290
865
0.127927
0.784191
76.773239
0.885480
866
0.123134
0.729571
60.989555
0.853589
867
0.119733
0.769269
69.834816
0.876822
868
0.123930
0.803790
75.501091
0.896240
869
0.123547
0.719518
61.469032
0.848037
870
0.128810
0.721589
67.405113
0.849203
871
0.125005
0.697703
67.956528
0.835003
872
0.122462
0.813084
73.111679
0.901506
873
0.122447
0.740543
73.670097
0.860176
874
0.126857
0.749528
72.324921
0.865690
875
0.131695
0.722813
61.078850
0.849694
876
0.125880
0.662522
62.707043
0.813879
877
0.126684
0.751525
70.285187
0.866694
878
0.124539
0.704868
59.220436
0.838918
879
0.124954
0.828557
72.955673
0.909943
880
0.123608
0.787326
71.788651
0.887207
881
0.127098
0.748183
60.022919
0.864737
882
0.126865
0.851146
87.225677
0.921999
883
0.125689
0.824449
82.027603
0.907264
884
0.124584
0.760740
68.316032
0.871976
885
0.120108
0.774543
61.065792
0.879950
886
0.125730
0.932809
85.708900
0.965033
887
0.124557
0.814857
73.208153
0.902515
888
0.123117
0.705696
68.562347
0.839692
889
0.121852
0.721715
61.536713
0.849473
890
0.126098
0.742216
60.987450
0.860817
891
0.120653
1.062375
103.519363
1.030538
892
0.123961
0.790269
77.207947
0.888449
893
0.120068
0.719687
70.695168
0.847840
894
0.119714
0.791245
75.368866
0.889199
895
0.119573
0.818452
73.549835
0.904163
896
0.122320
0.808618
78.424637
0.899149
897
0.125887
0.824732
80.290741
0.907920
898
0.117799
0.836124
85.060043
0.913563
899
0.121143
0.758994
70.913597
0.870981
900
0.125233
0.873077
83.457085
0.934128
901
0.120287
0.849922
80.327820
0.921709
902
0.122525
0.783704
79.133583
0.884678
903
0.123662
0.756394
67.055534
0.869606
904
0.121717
0.865878
70.123909
0.930122
905
0.120713
0.793093
85.052780
0.889698
906
0.121080
0.889493
85.164940
0.942971
907
0.117612
0.844185
80.960449
0.918127
908
0.120263
0.838418
78.550041
0.915385
909
0.119573
0.782555
78.009926
0.884186
910
0.120953
0.863941
78.142128
0.928819
911
0.121647
0.796701
68.384628
0.892066
912
0.122894
0.756591
75.006775
0.869502
913
0.118885
0.750115
70.856598
0.865811
914
0.117796
0.668032
67.041161
0.817016
915
0.122959
0.727480
66.006737
0.852831
916
0.121519
0.816069
79.551155
0.903176
917
0.120895
0.802148
79.725113
0.895292
918
0.122331
0.731320
71.896248
0.854683
919
0.124202
0.765659
71.717850
0.874894
920
0.117634
0.789400
74.646309
0.888323
921
0.121598
0.823749
77.870102
0.907297
922
0.119266
0.689247
56.827957
0.829969
923
0.118664
0.691824
62.945202
0.831548
924
0.121882
0.861561
68.371605
0.928117
925
0.121415
0.776154
65.963669
0.880535
926
0.116793
0.672379
63.294083
0.819957
927
0.117897
0.714725
66.252090
0.845114
928
0.124026
0.665319
62.029869
0.815544
929
0.118336
0.771087
73.835213
0.877774
930
0.119793
0.622166
60.023930
0.788672
931
0.117317
0.773960
68.661530
0.879605
932
0.114125
0.695153
67.658638
0.833332
933
0.115104
0.664740
63.792461
0.814830
934
0.116095
0.651460
59.220108
0.806791
935
0.117770
0.726300
67.102173
0.852068
936
0.119095
0.751121
69.539070
0.866558
937
0.118085
0.841159
76.659012
0.916974
938
0.117928
0.780033
73.175438
0.882683
939
0.118104
0.685101
69.956955
0.827517
940
0.120478
0.771169
76.381409
0.877659
941
0.117634
0.717389
64.905418
0.846399
942
0.114946
0.748336
65.818687
0.864469
943
0.118481
0.811096
73.063972
0.900270
944
0.119836
0.693087
67.192596
0.832341
945
0.117801
0.709374
69.899551
0.841849
946
0.118832
0.657450
56.023426
0.810399
947
0.113406
0.735711
60.315777
0.857450
948
0.115026
0.741877
63.647175
0.860206
949
0.114263
0.667443
56.309727
0.816779
950
0.118397
0.632594
57.012901
0.795348
951
0.111560
0.719670
66.968857
0.847906
952
0.114859
0.669670
63.329250
0.818244
953
0.119110
0.671627
61.439075
0.819363
954
0.116624
0.675068
64.733635
0.821046
955
0.115915
0.764605
76.854065
0.874040
956
0.121648
0.687593
65.689209
0.829128
957
0.119421
0.697026
59.113720
0.834478
958
0.115595
0.706061
72.140297
0.840119
959
0.116709
0.685859
55.077888
0.827981
960
0.116936
0.703268
57.799576
0.838525
961
0.116535
0.711215
60.594662
0.842464
962
0.115108
0.724686
68.987823
0.850802
963
0.116769
0.661107
64.263542
0.812487
964
0.113600
0.703931
67.577583
0.838645
965
0.113771
0.661129
64.234398
0.812776
966
0.115027
0.697155
69.484673
0.834809
967
0.114927
0.779625
69.883339
0.882858
968
0.114793
0.673794
59.328445
0.820541
969
0.113556
0.696457
59.943825
0.834325
970
0.115090
0.705856
67.017387
0.840031
971
0.115081
0.703994
63.391975
0.838366
972
0.113798
0.673404
61.300282
0.820445
973
0.115016
0.703637
66.659164
0.838355
974
0.114694
0.614552
55.502102
0.783201
975
0.115214
0.701869
62.894051
0.837469
976
0.112756
0.657801
61.919682
0.810937
977
0.113553
0.689850
64.984772
0.830112
978
0.114204
0.715862
60.413063
0.845453
979
0.116184
0.730430
62.707680
0.853831
980
0.114757
0.703285
62.496994
0.838057
981
0.113578
0.663716
60.079155
0.814641
982
0.116264
0.668989
62.524399
0.817625
983
0.114423
0.663821
60.894039
0.814687
984
0.110721
0.747818
71.188477
0.864558
985
0.111381
0.807214
66.767303
0.898053
986
0.116541
0.664790
60.660988
0.815089
987
0.113688
0.686736
62.330536
0.828502
988
0.112749
0.665166
58.630375
0.815163
989
0.112212
0.689947
58.273819
0.830407
990
0.112025
0.719338
69.021492
0.847998
991
0.113327
0.651904
66.724060
0.807113
992
0.110513
0.674112
62.859386
0.820740
993
0.112400
0.727665
74.631126
0.852418
994
0.112986
0.664349
62.482075
0.814801
995
0.115632
0.718697
62.276108
0.847678
996
0.111128
0.719331
64.033096
0.847973
997
0.113917
0.721044
63.799362
0.848813
998
0.109593
0.696601
67.893372
0.834553
999
0.112950
0.680917
62.757217
0.824996
1000
0.111084
0.692478
60.775936
0.832038
1001
0.114513
0.695487
66.445297
0.833748
1002
0.113680
0.647071
55.280186
0.803783
1003
0.108269
0.716334
74.150452
0.846165
1004
0.110694
0.724959
66.443893
0.851275
1005
0.113870
0.728502
70.775101
0.853326
1006
0.108967
0.716323
64.946159
0.845700
1007
0.113673
0.661401
63.587795
0.813233
1008
0.113013
0.638693
61.157848
0.798917
1009
0.115533
0.626402
60.000439
0.790945
1010
0.112777
0.639446
58.681961
0.799542
1011
0.113420
0.745895
71.443947
0.863412
1012
0.115846
0.697263
66.612900
0.834739
1013
0.110823
0.670315
66.377922
0.818662
1014
0.109037
0.732495
66.937401
0.855548
1015
0.113066
0.669594
58.175838
0.818005
1016
0.113751
0.712269
61.924576
0.843643
1017
0.113364
0.716702
66.051178
0.846051
1018
0.110154
0.629613
53.948437
0.793282
1019
0.113974
0.621548
55.322048
0.788056
1020
0.110751
0.601124
52.246807
0.775088
1021
0.109951
0.688833
67.933411
0.829391
1022
0.110413
0.662017
60.156349
0.813275
1023
0.106534
0.595584
55.279549
0.771388
1024
0.110037
0.651853
54.517361
0.807061
1025
0.108328
0.648169
61.931644
0.804919
1026
0.109239
0.707837
67.050537
0.840987
1027
0.110592
0.696830
65.411705
0.834121
1028
0.109364
0.616449
57.367199
0.784916
1029
0.108515
0.650146
60.075970
0.806090
1030
0.108698
0.626738
56.584175
0.791389
1031
0.110545
0.661287
57.550724
0.812886
1032
0.109648
0.716472
67.748405
0.846017
1033
0.110666
0.721703
61.370411
0.849336
1034
0.108052
0.702248
66.118401
0.837387
1035
0.110191
0.675877
60.395775
0.822072
1036
0.110227
0.748638
70.651085
0.864934
1037
0.107478
0.702911
60.390820
0.838160
1038
0.107014
0.682117
64.727257
0.825628
1039
0.107988
0.690681
59.398838
0.830663
1040
0.109406
0.726103
68.107109
0.851721
1041
0.111461
0.713802
67.855278
0.843992
1042
0.110008
0.738403
68.018517
0.859139
1043
0.111682
0.695826
61.646255
0.833998
1044
0.110149
0.707236
63.210236
0.840090
1045
0.110188
0.673537
61.170101
0.820452
1046
0.106136
0.675906
60.163101
0.821846
1047
0.110024
0.675090
63.409237
0.821519
1048
0.111450
0.694755
69.518837
0.833205
1049
0.109484
0.693412
68.181328
0.832619
1050
0.105962
0.723172
71.092850
0.850108
1051
0.107035
0.717332
68.434723
0.846860
1052
0.106150
0.690129
62.086918
0.830390
1053
0.107229
0.700486
67.600021
0.836539
1054
0.106542
0.700408
66.504951
0.836715
1055
0.105653
0.699423
63.906570
0.836009
1056
0.109556
0.693279
65.200073
0.832308
1057
0.110989
0.643094
60.262108
0.801174
1058
0.107899
0.678355
61.911251
0.823404
1059
0.108437
0.657593
57.979900
0.810708
1060
0.111720
0.623135
57.255482
0.788568
1061
0.105845
0.673802
61.635654
0.820593
1062
0.105884
0.663994
59.094601
0.814784
1063
0.106755
0.631936
55.873631
0.794906
1064
0.109200
0.683541
59.107258
0.826329
1065
0.108799
0.697166
60.864975
0.834405
1066
0.108162
0.679611
58.756157
0.824043
1067
0.106922
0.693146
58.279892
0.831820
1068
0.108234
0.661890
58.878780
0.813535
1069
0.108920
0.647930
56.688477
0.804521
1070
0.109097
0.673447
62.437794
0.820384
1071
0.109302
0.659194
58.287655
0.811860
1072
0.109136
0.677154
62.214500
0.822631
1073
0.106175
0.687666
62.000080
0.828926
1074
0.108152
0.676504
61.541306
0.822206
1075
0.109337
0.693769
62.072330
0.832747
1076
0.104329
0.660901
59.947762
0.812639
1077
0.105230
0.732236
67.541862
0.855634
1078
0.105147
0.677211
61.423275
0.822420
1079
0.106023
0.673557
59.437977
0.820588
1080
0.104371
0.679835
60.936131
0.824433
1081
0.106817
0.643510
56.378914
0.801805
1082
0.105415
0.629687
57.273369
0.793436
1083
0.107084
0.687497
63.435608
0.828780
1084
0.107208
0.675498
62.246037
0.821577
1085
0.106632
0.704333
62.489525
0.838713
1086
0.103562
0.712755
63.820599
0.843980
1087
0.106870
0.718676
64.974655
0.847621
1088
0.105173
0.637784
56.179050
0.798471
1089
0.104018
0.689378
60.056961
0.829875
1090
0.107031
0.710522
62.294357
0.842544
1091
0.105412
0.687441
63.063351
0.828904
1092
0.105692
0.670642
61.048313
0.818848
1093
0.102850
0.640987
60.003826
0.800379
1094
0.107203
0.683166
63.222614
0.825690
1095
0.104189
0.695009
67.003403
0.833569
1096
0.103466
0.675452
65.622917
0.821270
1097
0.105225
0.724957
71.218498
0.851078
1098
0.105729
0.699567
65.197617
0.835857
1099
0.103115
0.663635
61.898689
0.814576
1100
0.102816
0.685389
62.144707
0.827651
1101
0.105137
0.669069
62.819530
0.817756
1102
0.105772
0.674110
63.492130
0.820673
1103
0.107906
0.658794
57.200062
0.810734
1104
0.105887
0.674157
60.639843
0.820823
1105
0.105918
0.658582
58.905724
0.811026
1106
0.105836
0.649627
58.043644
0.805798
1107
0.105568
0.714683
70.111183
0.845331
1108
0.105678
0.720048
68.355591
0.848493
1109
0.104799
0.673262
61.219707
0.820322
1110
0.104972
0.662744
62.762913
0.813806
1111
0.105133
0.690532
61.640652
0.830503
1112
0.103486
0.664740
63.140549
0.814835
1113
0.101623
0.684902
64.553047
0.827519
1114
0.103549
0.663503
59.322552
0.814103
1115
0.102538
0.682376
60.343124
0.825887
1116
0.104272
0.681480
60.307819
0.825448
1117
0.103168
0.644983
52.785774
0.802900
1118
0.104473
0.680538
61.959587
0.824433
1119
0.103827
0.683811
66.384262
0.826732
1120
0.105846
0.646476
58.515076
0.803625
1121
0.105947
0.659887
59.862263
0.812283
1122
0.104574
0.672116
60.500851
0.819222
1123
0.103727
0.690917
62.123569
0.830868
1124
0.105882
0.695593
60.763470
0.833798
1125
0.100477
0.685381
58.069469
0.827735
1126
0.100181
0.658140
57.167770
0.810719
1127
0.104611
0.674485
57.494442
0.821196
1128
0.103275
0.690486
61.308189
0.830830
1129
0.104447
0.679717
60.872749
0.824328
1130
0.105742
0.677272
58.419724
0.822878
1131
0.105764
0.672718
61.970825
0.820070
1132
0.106349
0.680749
60.648876
0.824813
1133
0.104641
0.658563
56.587170
0.811158
1134
0.103383
0.663976
54.905663
0.814585
1135
0.101461
0.670168
55.388039
0.818291
1136
0.102809
0.663206
60.060989
0.813903
1137
0.103380
0.663598
56.374207
0.814369
1138
0.101517
0.656552
57.405998
0.810118
1139
0.103785
0.683580
61.629467
0.826476
1140
0.104309
0.654732
56.688705
0.808792
1141
0.104922
0.667449
59.425304
0.816854
1142
0.102536
0.693071
67.670326
0.832449
1143
0.100758
0.646690
54.260036
0.803960
1144
0.105538
0.657214
57.449348
0.810258
1145
0.101753
0.665163
56.339130
0.815167
1146
0.103004
0.645178
56.246162
0.803142
1147
0.102237
0.668446
57.747051
0.817382
1148
0.102180
0.665866
59.391396
0.815635
1149
0.105951
0.654319
54.368999
0.808872
1150
0.103076
0.655487
57.211899
0.808585
1151
0.099158
0.682879
62.356533
0.826283
1152
0.101845
0.675864
59.867569
0.821959
1153
0.101236
0.672338
59.207954
0.819734
1154
0.104897
0.663111
54.334721
0.814174
1155
0.103444
0.655755
56.016613
0.809094
1156
0.102822
0.662476
56.886436
0.813638
1157
0.103356
0.663183
58.463024
0.814197
1158
0.102910
0.672716
58.736458
0.820122
1159
0.103234
0.679339
58.408337
0.823705
1160
0.103107
0.678410
60.128101
0.823387
1161
0.101940
0.668796
60.847252
0.817480
1162
0.103096
0.650176
55.445995
0.806185
1163
0.103919
0.639399
54.183998
0.799372
1164
0.104967
0.668196
57.115814
0.817269
1165
0.100913
0.661402
57.875488
0.812575
1166
0.100585
0.666914
56.899414
0.815597
1167
0.103432
0.642739
54.003281
0.801539
1168
0.100148
0.645571
53.682407
0.803263
1169
0.103059
0.655833
58.303867
0.809653
1170
0.105150
0.649880
55.440018
0.805197
1171
0.103878
0.680289
60.441952
0.824607
1172
0.100066
0.688987
62.761875
0.830011
1173
0.100683
0.643519
54.636375
0.801750
1174
0.103695
0.662101
56.798450
0.813337
1175
0.101431
0.654249
56.248837
0.808412
1176
0.100923
0.659488
56.069420
0.811805
1177
0.100472
0.668836
58.301693
0.817303
1178
0.104858
0.706710
64.259979
0.840536
1179
0.105970
0.665121
57.154869
0.815120
1180
0.103628
0.660857
57.652969
0.812251
1181
0.098558
0.648798
55.659714
0.805202
1182
0.100379
0.660021
56.983532
0.812224
1183
0.102600
0.662823
57.869301
0.813742
1184
0.100344
0.646766
54.533501
0.803965
1185
0.102849
0.646764
52.892136
0.803796
1186
0.103438
0.668249
58.565964
0.817379
1187
0.102145
0.668463
58.578476
0.817300
1188
0.103654
0.672041
58.621674
0.819339
1189
0.101864
0.643152
54.239857
0.801821
1190
0.101938
0.636780
51.671230
0.797895
1191
0.102562
0.666655
58.351452
0.816230
1192
0.103810
0.649241
55.872070
0.805692
1193
0.101447
0.653841
53.782349
0.808504
1194
0.101819
0.654426
55.459511
0.808151
1195
0.101880
0.662153
56.282032
0.813545
1196
0.099991
0.653965
55.536018
0.808547
1197
0.102185
0.632493
51.019947
0.794836
1198
0.103757
0.672557
58.831612
0.819726
1199
0.103230
0.644949
54.025200
0.802998
1200
0.101254
0.668392
58.039326
0.817496
In [31]:
l.recorder.plot_losses()
In [6]:
l.save(f"r_speedup_{optimizer}_batch_norm_{batch_norm}_{loss_func}_nlayers_{len(layers_sizes)}_log_{log}")
In [7]:
!ls models
old_models
old_repr
r_speedup_Adam_batch_norm_True_MAPE_nlayers_5_log_False.pth
speedup_Adam_batch_norm_True_MAPE_nlayers_5_log_False2.pth
speedup_Adam_batch_norm_True_MAPE_nlayers_5_log_False.pth
speedup_Adam_batch_norm_True_MSE_nlayers_5_log_False.pth
speedup_Adam_batch_norm_True_MSE_nlayers_5_log_True.pth
tmp.pth
In [12]:
val_df = get_results_df(val_dl, l.model)
train_df = get_results_df(train_dl, l.model)
In [13]:
df = train_df
In [14]:
df[:][['prediction','target', 'abs_diff','APE']].describe()
Out[14]:
prediction
target
abs_diff
APE
count
245283.000000
245283.000000
2.452830e+05
245283.000000
mean
1.102805
1.135724
2.135267e-01
22.577551
std
1.295271
1.405682
5.665563e-01
91.278419
min
0.010044
0.008491
2.048910e-07
0.000030
25%
0.280880
0.278690
5.817745e-03
1.646686
50%
0.855459
0.899071
3.095371e-02
5.807508
75%
1.050393
1.036481
1.352153e-01
17.479273
max
8.452207
16.089287
1.541526e+01
5824.463867
In [15]:
df = val_df
In [16]:
df[:][['prediction','target', 'abs_diff','APE']].describe()
Out[16]:
prediction
target
abs_diff
APE
count
10000.000000
10000.000000
10000.000000
10000.000000
mean
1.176222
1.430897
0.442006
42.532043
std
1.231279
1.683147
0.734061
148.393097
min
0.014810
0.014795
0.000040
0.006628
25%
0.389919
0.395193
0.024651
4.690779
50%
0.961758
1.000000
0.134899
22.480320
75%
1.257520
1.621683
0.497272
48.398965
max
7.889497
10.872228
6.232839
5399.246094
In [35]:
df[:][['index','name','prediction','target', 'abs_diff','APE']].to_csv(path_or_buf='./eval_results.csv',sep=';')
In [18]:
df[(df.interchange==0) & (df.unroll == 0) & (df.tile == 0)][['prediction','target', 'abs_diff','APE']].describe()
Out[18]:
prediction
target
abs_diff
APE
count
2074.000000
2074.0
2074.000000
2074.000000
mean
0.998558
1.0
0.050076
5.007599
std
0.208153
0.0
0.202042
20.204212
min
0.603882
1.0
0.000066
0.006628
25%
0.979231
1.0
0.002551
0.255150
50%
0.995987
1.0
0.004715
0.471514
75%
0.997637
1.0
0.022805
2.280509
max
6.001675
1.0
5.001675
500.167511
In [19]:
df[(df.interchange==0) & (df.unroll == 0) & (df.tile == 1)][['prediction','target', 'abs_diff','APE']].describe()
Out[19]:
prediction
target
abs_diff
APE
count
725.000000
725.000000
725.000000
725.000000
mean
1.462764
1.519225
0.393448
29.481096
std
1.346193
1.422773
0.573493
31.370106
min
0.145803
0.091200
0.000040
0.009375
25%
0.494025
0.508859
0.070696
8.597638
50%
0.991690
0.993141
0.191803
19.992519
75%
2.151304
2.207209
0.496262
41.369717
max
7.768654
9.358214
3.835224
295.724640
In [21]:
df[(df.interchange==0) & (df.unroll == 1) & (df.tile == 0)][['prediction','target', 'abs_diff','APE']].describe()
Out[21]:
prediction
target
abs_diff
APE
count
281.000000
281.000000
281.000000
281.000000
mean
4.372236
6.044158
1.848176
66.339111
std
2.171317
3.221366
1.231329
163.969162
min
0.904279
0.105179
0.001005
0.060693
25%
1.006489
2.546717
1.032713
18.170967
50%
5.899143
7.389543
1.563098
31.399256
75%
5.972240
8.668690
2.789691
43.133614
max
6.070222
10.872228
4.899988
848.387634
In [22]:
df[(df.interchange==1) & (df.unroll == 0) & (df.tile == 0)][['prediction','target', 'abs_diff','APE']].describe()
Out[22]:
prediction
target
abs_diff
APE
count
232.000000
232.000000
232.000000
232.000000
mean
0.913173
0.918692
0.347373
122.910507
std
1.018806
1.168210
0.588294
610.275146
min
0.084280
0.018092
0.001015
0.220132
25%
0.260828
0.265846
0.033345
7.204351
50%
0.456005
0.466655
0.101908
19.017271
75%
1.092822
0.956079
0.327444
41.838287
max
5.042413
8.069739
3.409136
5399.246094
In [23]:
df[(df.interchange==0) & (df.unroll == 1) & (df.tile == 1)][['prediction','target', 'abs_diff','APE']].describe()
Out[23]:
prediction
target
abs_diff
APE
count
868.000000
868.000000
868.000000
868.000000
mean
1.589693
1.748770
0.590289
37.538128
std
1.585954
1.605983
0.793788
50.443039
min
0.168853
0.057130
0.000085
0.043231
25%
0.442154
0.695116
0.105190
12.070397
50%
1.012416
1.241629
0.302328
29.301636
75%
2.136123
2.229650
0.773482
44.885977
max
7.889497
10.137201
4.684408
414.385712
In [24]:
df[(df.interchange==1) & (df.unroll == 1) & (df.tile == 0)][['prediction','target', 'abs_diff','APE']].describe()
Out[24]:
prediction
target
abs_diff
APE
count
1276.000000
1276.000000
1276.000000
1276.000000
mean
1.962418
2.878316
1.036611
72.382378
std
1.334146
2.108479
1.021500
274.414642
min
0.101093
0.042484
0.000331
0.041299
25%
0.819918
0.977715
0.154067
10.751699
50%
2.072164
2.629361
0.717993
40.390493
75%
2.634203
4.365878
1.818399
50.273042
max
4.919624
9.784180
5.172597
2245.031982
In [25]:
df[(df.interchange==1) & (df.unroll == 0) & (df.tile == 1)][['prediction','target', 'abs_diff','APE']].describe()
Out[25]:
prediction
target
abs_diff
APE
count
1663.000000
1663.000000
1663.000000
1663.000000
mean
0.827254
0.937209
0.295498
32.696384
std
0.851419
1.023356
0.457449
32.905022
min
0.014810
0.014836
0.000083
0.030472
25%
0.217906
0.246893
0.038760
9.840522
50%
0.529727
0.624512
0.132425
24.348541
75%
1.120389
1.173670
0.354944
48.979925
max
4.983345
8.724252
5.141850
309.855896
In [26]:
df[(df.interchange==1) & (df.unroll == 1) & (df.tile == 1)][['prediction','target', 'abs_diff','APE']].describe()
Out[26]:
prediction
target
abs_diff
APE
count
2881.000000
2881.000000
2881.000000
2881.000000
mean
0.670118
0.858291
0.383388
57.996346
std
0.757664
1.094979
0.572025
79.790131
min
0.016312
0.014795
0.000374
0.131162
25%
0.158396
0.183774
0.059443
19.831461
50%
0.420433
0.424563
0.154730
43.728092
75%
0.847958
1.033311
0.456153
58.873505
max
4.961184
9.207530
6.232839
1148.485840
In [27]:
df[(df.interchange + df.tile + df.unroll != 0)][['prediction','target', 'abs_diff','APE']].describe()
Out[27]:
prediction
target
abs_diff
APE
count
7926.000000
7926.000000
7926.000000
7926.000000
mean
1.222709
1.543651
0.544563
52.351154
std
1.375149
1.874331
0.786424
164.959198
min
0.014810
0.014795
0.000040
0.009375
25%
0.304365
0.312547
0.065765
12.863225
50%
0.714742
0.830516
0.205677
34.351240
75%
1.788819
2.132191
0.660448
52.750795
max
7.889497
10.872228
6.232839
5399.246094
In [28]:
df1 = df[(df.interchange==0) & (df.unroll == 0) & (df.tile == 0)]
joint_plot(df1, f"Validation dataset, {loss_func} loss")
df1 = df[(df.interchange==0) & (df.unroll == 0) & (df.tile == 1)]
joint_plot(df1, f"Validation dataset, {loss_func} loss")
df1 = df[(df.interchange==0) & (df.unroll == 1) & (df.tile == 0)]
joint_plot(df1, f"Validation dataset, {loss_func} loss")
df1 = df[(df.interchange==1) & (df.unroll == 0) & (df.tile == 0)]
joint_plot(df1, f"Validation dataset, {loss_func} loss")
df1 = df[(df.interchange==0) & (df.unroll == 1) & (df.tile == 1)]
joint_plot(df1, f"Validation dataset, {loss_func} loss")
df1 = df[(df.interchange==1) & (df.unroll == 1) & (df.tile == 0)]
joint_plot(df1, f"Validation dataset, {loss_func} loss")
df1 = df[(df.interchange==1) & (df.unroll == 0) & (df.tile == 1)]
joint_plot(df1, f"Validation dataset, {loss_func} loss")
df1 = df[(df.interchange==0) & (df.unroll == 1) & (df.tile == 1)]
joint_plot(df1, f"Validation dataset, {loss_func} loss")
df1 = df[(df.interchange==1) & (df.unroll == 1) & (df.tile == 1)]
joint_plot(df1, f"Validation dataset, {loss_func} loss")
df1 = df[(df.interchange + df.tile + df.unroll != 0)]
joint_plot(df1, f"Validation dataset, {loss_func} loss")
df2 = df
joint_plot(df2, f"Validation dataset, {loss_func} loss")
/data/scratch/henni-mohammed/anaconda3/lib/python3.7/site-packages/scipy/stats/stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.
return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval
In [ ]:
In [ ]:
Content source: rbaghdadi/COLi
Similar notebooks: