I realized I made a mistake in the design of the Gendy GK module: the module is based on the premise that all arguments to a synth are real valued. The Gendy parameters ampdist and durdist are categorical variables and as such violate the assumption.
In [1]:
import sys
sys.path.append('/Users/spacecoffin/Development')
import GravelKicker as gk
import librosa
import numpy as np
import os
import pandas as pd
from datetime import datetime
from supriya.tools import nonrealtimetools
In [2]:
this_dir = '/Users/spacecoffin/Development/GravelKicker/__gen_files'
In [6]:
pmtx = gk.generator.gendy1.gen_params(rows=100)
In [7]:
df = gk.generator.gendy1.format_params(pmtx)
In [73]:
df.to_pickle()
Out[73]:
adparam
ampscale
ddparam
durscale
ampdist
durdist
knum
minfrequency
maxfrequency
init_cps
hash
0
0.140567
0.296349
0.694119
0.870140
5.0
5.0
12.0
27.500000
123.470825
16.0
bffb8f3b4b4cb178cb74ec7943c7eb44
1
0.976231
0.277556
0.617027
0.173149
5.0
5.0
11.0
16.351598
25.956544
16.0
ca190a8ae0e5c4beb7d8e6b38bfbf1e0
2
0.639036
0.779432
0.954356
0.248040
5.0
0.0
14.0
220.000000
932.327523
16.0
7683a96287dbca2c7a7f870a293c459a
3
0.856778
0.615598
0.114462
0.865524
2.0
2.0
10.0
415.304698
1975.533205
16.0
4745a743ac821963e58c2aa039d87647
4
0.717413
0.886676
0.446664
0.429450
0.0
5.0
14.0
10.300861
34.647829
16.0
7b0b9271b3717641f77e707512cbce6b
5
0.726739
0.447664
0.455060
0.594067
3.0
4.0
11.0
25.956544
277.182631
16.0
c5bd8e39442f83b56f2cdf371bac4dcc
6
0.315652
0.150260
0.696117
0.245300
5.0
1.0
12.0
58.270470
220.000000
16.0
c44badc8e9e959796573b5ab746207e9
7
0.012038
0.055573
0.436362
0.435634
2.0
3.0
13.0
233.081881
293.664768
16.0
a395dfbc10e3c95ef5b2b476dabe4291
8
0.434207
0.600596
0.144068
0.845757
0.0
2.0
11.0
61.735413
73.416192
16.0
475b09a4257f6f72ccc8bbd999e5bf97
9
0.314423
0.720786
0.898457
0.501920
0.0
5.0
12.0
46.249303
87.307058
16.0
d6eba37b4e5b9c45f8fa0e57e52d925d
10
0.149213
0.803590
0.506268
0.221940
3.0
5.0
14.0
14.567618
27.500000
16.0
2951fd9b5a76328b27f75feb84c63397
11
0.941576
0.793607
0.884417
0.507369
5.0
4.0
13.0
92.498606
587.329536
16.0
9a3a90d01ef5d5322fe768e04d66c291
12
0.847062
0.948210
0.166209
0.483142
2.0
3.0
9.0
233.081881
391.995436
16.0
85f21a559d08acf0d6b0760d9b65b4c0
13
0.774483
0.515727
0.191346
0.067267
3.0
1.0
13.0
24.499715
82.406889
16.0
f10c90377d5ae11ce4c38e398a5cdc09
14
0.387323
0.828988
0.305139
0.462204
1.0
2.0
13.0
41.203445
311.126984
16.0
2c0bd3f2742a34e9cfdb567082cbad4a
15
0.264970
0.146887
0.802805
0.599800
5.0
3.0
14.0
184.997211
554.365262
16.0
ca3b4c649f3d657da8a1de6c234d538c
16
0.398866
0.772135
0.224419
0.436501
4.0
3.0
12.0
220.000000
1567.981744
16.0
ea524ef020271fa6572fccaba4196a74
17
0.374543
0.151519
0.104026
0.385049
2.0
1.0
15.0
311.126984
1567.981744
16.0
d83814cf8b284700f3f8c0605001c4d8
18
0.406214
0.523283
0.637029
0.515608
4.0
0.0
12.0
103.826174
123.470825
16.0
8eda513dc4b14111ce82b8227b585c6a
19
0.025347
0.748021
0.571764
0.818625
3.0
1.0
10.0
233.081881
369.994423
16.0
22080531ee67adb3cc6c0ecf3dd0de2b
20
0.630945
0.341776
0.907955
0.846463
1.0
1.0
13.0
554.365262
783.990872
16.0
dde1914258e216d2a18e4570dc9dcf4b
21
0.549380
0.549365
0.514414
0.067968
4.0
3.0
11.0
29.135235
466.163762
16.0
56c44a92179bac1efef143ffebf55d75
22
0.624507
0.970800
0.115833
0.084680
1.0
4.0
9.0
10.300861
21.826764
16.0
a3296cf59751fa190ec522c6f87be33c
23
0.619147
0.702963
0.704624
0.846077
1.0
2.0
12.0
391.995436
415.304698
16.0
5e85cdf07830af641cc2ec47d952fafc
24
0.441448
0.929197
0.923300
0.362459
4.0
0.0
12.0
9.177024
36.708096
16.0
60c155ef9e92536d40fbc7f7b7157864
25
0.049221
0.096192
0.506798
0.949034
5.0
1.0
13.0
184.997211
246.941651
16.0
50ca88648059b1d4694978e165dcaf3f
26
0.868060
0.107455
0.031129
0.227593
2.0
0.0
13.0
466.163762
739.988845
16.0
00863d513033675d19b28640fd0d395f
27
0.438835
0.668790
0.221935
0.199449
5.0
5.0
12.0
38.890873
61.735413
16.0
6c52b9816e7b197e86dacfeef5ec8411
28
0.544936
0.117482
0.435654
0.900235
2.0
0.0
10.0
30.867706
2793.825851
16.0
94f4ac6db9370e674e5a9a895f2452be
29
0.395026
0.683075
0.127904
0.695616
3.0
1.0
11.0
61.735413
369.994423
16.0
d33d878df94648823c074a6dd23b3eb5
...
...
...
...
...
...
...
...
...
...
...
...
70
0.633824
0.100791
0.886926
0.569303
3.0
3.0
10.0
195.997718
207.652349
16.0
134d83ae08471458100e2b29d92b4641
71
0.483271
0.043196
0.889111
0.030375
5.0
1.0
11.0
61.735413
87.307058
16.0
d52e1bd1aa27cf25b8aef8872d2ed51a
72
0.487946
0.690625
0.478157
0.042380
1.0
4.0
8.0
43.653529
184.997211
16.0
e9b73567a5a5671a8eb044e60fd33a90
73
0.041706
0.854476
0.204663
0.241624
1.0
3.0
9.0
1108.730524
2093.004522
16.0
9c1883ecedb173ef8c7ee4a829c09a41
74
0.593181
0.204258
0.625244
0.102961
1.0
5.0
12.0
311.126984
2217.461048
16.0
fc6585b0e0b3571c8769ba7b7346ccbf
75
0.775826
0.284496
0.859170
0.631275
2.0
2.0
11.0
9.177024
10.913382
16.0
67e3d72503af4c732a564037deb4fa5c
76
0.108605
0.609662
0.596985
0.349631
2.0
4.0
12.0
130.812783
246.941651
16.0
d9c9fdced4def3bd9a76a1481badf29d
77
0.120932
0.681908
0.921700
0.167986
3.0
0.0
12.0
8.175799
4186.009045
16.0
0fb3d4bc4af8cb28f9b4f37345ea3dd8
78
0.943481
0.172716
0.731616
0.235319
5.0
3.0
8.0
164.813778
329.627557
16.0
49397f5d04dc31a874ee579451095f91
79
0.336802
0.893346
0.556808
0.423930
4.0
0.0
8.0
38.890873
3951.066410
16.0
7f2d453125256b5939108f3999d9b5b1
80
0.332562
0.432787
0.198116
0.004707
2.0
4.0
13.0
220.000000
3951.066410
16.0
d2256c5075e82c838e0e0b586b627caf
81
0.774138
0.324171
0.095434
0.680617
5.0
5.0
12.0
1108.730524
3951.066410
16.0
8eb6744d5196cd79fdb1ab0d0ae9e469
82
0.200390
0.480073
0.203515
0.801010
1.0
0.0
14.0
466.163762
622.253967
16.0
b16591eb1f7da5e8268f6ed167b2f195
83
0.388212
0.723324
0.908967
0.115439
5.0
2.0
11.0
65.406391
87.307058
16.0
22cd678b90d3743a8c6e87d7c5b5468c
84
0.020576
0.584655
0.636087
0.277855
2.0
3.0
11.0
195.997718
293.664768
16.0
1df28e5560ce0d633abdb0e099bd5d8e
85
0.492623
0.989929
0.976492
0.008921
2.0
1.0
14.0
587.329536
1864.655046
16.0
4764193186d0d93fb800d1d77b7cc9bb
86
0.759948
0.059714
0.033785
0.830425
4.0
0.0
12.0
97.998859
130.812783
16.0
54280d7e4b6824cbf515051646306315
87
0.233052
0.395877
0.780626
0.612113
0.0
4.0
12.0
38.890873
123.470825
16.0
bfaa65bfaa7747e6e21e863a2cd8364e
88
0.625776
0.173321
0.732628
0.232759
0.0
5.0
13.0
329.627557
783.990872
16.0
6f5dff1745e9ea02efd2f130dbcfcfcc
89
0.613349
0.470304
0.207821
0.158491
2.0
1.0
14.0
207.652349
246.941651
16.0
500c593134a3c56e70db47280b710693
90
0.588139
0.902430
0.652230
0.562310
0.0
5.0
11.0
1661.218790
3951.066410
16.0
c97ce82fd509c146e035bf63ac69ce37
91
0.232791
0.267941
0.800947
0.251281
4.0
3.0
13.0
38.890873
77.781746
16.0
0b236efc6c5b8a31c9c7c92feb94321e
92
0.068025
0.064293
0.086288
0.911258
0.0
2.0
11.0
97.998859
1864.655046
16.0
fa53d986b206f52b5b07ae47dbef2723
93
0.463363
0.789818
0.607833
0.110046
3.0
3.0
12.0
8.661957
13.750000
16.0
fd36521a7917cfbcb51946a31ef81347
94
0.044057
0.291824
0.094686
0.832175
4.0
1.0
11.0
1174.659072
1864.655046
16.0
910e1b3c69e42698e5c591ae94ef28fd
95
0.459862
0.349461
0.907124
0.839088
5.0
4.0
13.0
8.175799
65.406391
16.0
3609610b897255f366b3ac0f44cf1b3e
96
0.592219
0.465846
0.950628
0.814859
5.0
5.0
12.0
16.351598
82.406889
16.0
07a7a83f48846d4624d736efdebc2b85
97
0.301785
0.953625
0.682601
0.126688
2.0
5.0
11.0
61.735413
523.251131
16.0
36b984e6f2dc3b89fa810ce28de4d18f
98
0.387658
0.406595
0.797436
0.941026
0.0
0.0
9.0
116.540940
220.000000
16.0
904a6b2fb54c1cb1c218cc859e6092fc
99
0.817273
0.740969
0.804849
0.691434
3.0
2.0
10.0
34.647829
123.470825
16.0
d68fc0c924526c94a151671a261ac705
100 rows × 11 columns
In [30]:
pmtx
Out[30]:
array([[ 1.40566725e-01, 2.96348706e-01, 6.94119437e-01,
8.70139674e-01, 5.00000000e+00, 5.00000000e+00,
1.20000000e+01, 2.75000000e+01, 1.23470825e+02,
1.60000000e+01],
[ 9.76231054e-01, 2.77556017e-01, 6.17027301e-01,
1.73149196e-01, 5.00000000e+00, 5.00000000e+00,
1.10000000e+01, 1.63515978e+01, 2.59565436e+01,
1.60000000e+01],
[ 6.39036413e-01, 7.79432295e-01, 9.54356368e-01,
2.48040346e-01, 5.00000000e+00, 0.00000000e+00,
1.40000000e+01, 2.20000000e+02, 9.32327523e+02,
1.60000000e+01],
[ 8.56777788e-01, 6.15597540e-01, 1.14462311e-01,
8.65524238e-01, 2.00000000e+00, 2.00000000e+00,
1.00000000e+01, 4.15304698e+02, 1.97553321e+03,
1.60000000e+01],
[ 7.17413492e-01, 8.86676141e-01, 4.46664321e-01,
4.29449566e-01, 0.00000000e+00, 5.00000000e+00,
1.40000000e+01, 1.03008612e+01, 3.46478289e+01,
1.60000000e+01],
[ 7.26738693e-01, 4.47663964e-01, 4.55059651e-01,
5.94066943e-01, 3.00000000e+00, 4.00000000e+00,
1.10000000e+01, 2.59565436e+01, 2.77182631e+02,
1.60000000e+01],
[ 3.15652363e-01, 1.50259658e-01, 6.96117114e-01,
2.45300437e-01, 5.00000000e+00, 1.00000000e+00,
1.20000000e+01, 5.82704702e+01, 2.20000000e+02,
1.60000000e+01],
[ 1.20383750e-02, 5.55734758e-02, 4.36362136e-01,
4.35634401e-01, 2.00000000e+00, 3.00000000e+00,
1.30000000e+01, 2.33081881e+02, 2.93664768e+02,
1.60000000e+01],
[ 4.34207031e-01, 6.00596427e-01, 1.44068068e-01,
8.45757042e-01, 0.00000000e+00, 2.00000000e+00,
1.10000000e+01, 6.17354127e+01, 7.34161920e+01,
1.60000000e+01],
[ 3.14422876e-01, 7.20785614e-01, 8.98456932e-01,
5.01920043e-01, 0.00000000e+00, 5.00000000e+00,
1.20000000e+01, 4.62493028e+01, 8.73070579e+01,
1.60000000e+01],
[ 1.49212671e-01, 8.03590202e-01, 5.06268250e-01,
2.21940402e-01, 3.00000000e+00, 5.00000000e+00,
1.40000000e+01, 1.45676175e+01, 2.75000000e+01,
1.60000000e+01],
[ 9.41576313e-01, 7.93607339e-01, 8.84416819e-01,
5.07369355e-01, 5.00000000e+00, 4.00000000e+00,
1.30000000e+01, 9.24986057e+01, 5.87329536e+02,
1.60000000e+01],
[ 8.47061557e-01, 9.48209825e-01, 1.66209055e-01,
4.83142358e-01, 2.00000000e+00, 3.00000000e+00,
9.00000000e+00, 2.33081881e+02, 3.91995436e+02,
1.60000000e+01],
[ 7.74482971e-01, 5.15726527e-01, 1.91346006e-01,
6.72673587e-02, 3.00000000e+00, 1.00000000e+00,
1.30000000e+01, 2.44997147e+01, 8.24068892e+01,
1.60000000e+01],
[ 3.87323079e-01, 8.28987787e-01, 3.05138847e-01,
4.62203530e-01, 1.00000000e+00, 2.00000000e+00,
1.30000000e+01, 4.12034446e+01, 3.11126984e+02,
1.60000000e+01],
[ 2.64969927e-01, 1.46886565e-01, 8.02804672e-01,
5.99799814e-01, 5.00000000e+00, 3.00000000e+00,
1.40000000e+01, 1.84997211e+02, 5.54365262e+02,
1.60000000e+01],
[ 3.98865707e-01, 7.72134750e-01, 2.24418877e-01,
4.36500907e-01, 4.00000000e+00, 3.00000000e+00,
1.20000000e+01, 2.20000000e+02, 1.56798174e+03,
1.60000000e+01],
[ 3.74542711e-01, 1.51518953e-01, 1.04026171e-01,
3.85049362e-01, 2.00000000e+00, 1.00000000e+00,
1.50000000e+01, 3.11126984e+02, 1.56798174e+03,
1.60000000e+01],
[ 4.06214381e-01, 5.23283246e-01, 6.37029038e-01,
5.15607815e-01, 4.00000000e+00, 0.00000000e+00,
1.20000000e+01, 1.03826174e+02, 1.23470825e+02,
1.60000000e+01],
[ 2.53474277e-02, 7.48021172e-01, 5.71764387e-01,
8.18624821e-01, 3.00000000e+00, 1.00000000e+00,
1.00000000e+01, 2.33081881e+02, 3.69994423e+02,
1.60000000e+01],
[ 6.30944762e-01, 3.41775929e-01, 9.07954901e-01,
8.46462698e-01, 1.00000000e+00, 1.00000000e+00,
1.30000000e+01, 5.54365262e+02, 7.83990872e+02,
1.60000000e+01],
[ 5.49380016e-01, 5.49365418e-01, 5.14414410e-01,
6.79679725e-02, 4.00000000e+00, 3.00000000e+00,
1.10000000e+01, 2.91352351e+01, 4.66163762e+02,
1.60000000e+01],
[ 6.24506891e-01, 9.70799762e-01, 1.15832663e-01,
8.46796122e-02, 1.00000000e+00, 4.00000000e+00,
9.00000000e+00, 1.03008612e+01, 2.18267645e+01,
1.60000000e+01],
[ 6.19146520e-01, 7.02963089e-01, 7.04623715e-01,
8.46076739e-01, 1.00000000e+00, 2.00000000e+00,
1.20000000e+01, 3.91995436e+02, 4.15304698e+02,
1.60000000e+01],
[ 4.41448269e-01, 9.29196994e-01, 9.23300195e-01,
3.62458668e-01, 4.00000000e+00, 0.00000000e+00,
1.20000000e+01, 9.17702400e+00, 3.67080960e+01,
1.60000000e+01],
[ 4.92206884e-02, 9.61916363e-02, 5.06798484e-01,
9.49034410e-01, 5.00000000e+00, 1.00000000e+00,
1.30000000e+01, 1.84997211e+02, 2.46941651e+02,
1.60000000e+01],
[ 8.68059877e-01, 1.07454922e-01, 3.11292816e-02,
2.27593498e-01, 2.00000000e+00, 0.00000000e+00,
1.30000000e+01, 4.66163762e+02, 7.39988845e+02,
1.60000000e+01],
[ 4.38835281e-01, 6.68790457e-01, 2.21934772e-01,
1.99449462e-01, 5.00000000e+00, 5.00000000e+00,
1.20000000e+01, 3.88908730e+01, 6.17354127e+01,
1.60000000e+01],
[ 5.44935871e-01, 1.17482472e-01, 4.35653652e-01,
9.00235117e-01, 2.00000000e+00, 0.00000000e+00,
1.00000000e+01, 3.08677063e+01, 2.79382585e+03,
1.60000000e+01],
[ 3.95026275e-01, 6.83074667e-01, 1.27903847e-01,
6.95616148e-01, 3.00000000e+00, 1.00000000e+00,
1.10000000e+01, 6.17354127e+01, 3.69994423e+02,
1.60000000e+01],
[ 3.43855395e-01, 7.99278448e-01, 1.31273567e-01,
2.13649375e-01, 4.00000000e+00, 4.00000000e+00,
1.20000000e+01, 3.91995436e+02, 7.39988845e+02,
1.60000000e+01],
[ 6.60814016e-01, 4.66262544e-01, 8.11170039e-01,
8.29434882e-01, 0.00000000e+00, 2.00000000e+00,
1.40000000e+01, 2.61625565e+02, 3.11126984e+02,
1.60000000e+01],
[ 9.76757669e-01, 4.37239896e-01, 3.33418054e-01,
7.90673838e-01, 0.00000000e+00, 2.00000000e+00,
9.00000000e+00, 1.15623257e+01, 2.18267645e+01,
1.60000000e+01],
[ 3.46096712e-01, 6.65650199e-03, 8.66110898e-02,
9.70710505e-01, 3.00000000e+00, 3.00000000e+00,
1.30000000e+01, 1.74614116e+02, 2.46941651e+02,
1.60000000e+01],
[ 9.01814461e-02, 2.05660623e-01, 5.37871528e-01,
7.43020144e-02, 5.00000000e+00, 3.00000000e+00,
1.40000000e+01, 5.19130872e+01, 1.16540940e+02,
1.60000000e+01],
[ 6.96076211e-01, 4.14700477e-01, 2.43478701e-01,
5.35827853e-01, 5.00000000e+00, 1.00000000e+00,
1.20000000e+01, 2.44997147e+01, 1.30812783e+02,
1.60000000e+01],
[ 4.97783418e-01, 1.15200605e-01, 6.38931637e-01,
4.46775202e-01, 5.00000000e+00, 5.00000000e+00,
1.00000000e+01, 1.83540480e+01, 3.67080960e+01,
1.60000000e+01],
[ 9.16018992e-01, 3.72392766e-01, 4.64937396e-01,
3.73454475e-01, 1.00000000e+00, 1.00000000e+00,
1.10000000e+01, 1.66121879e+03, 2.21746105e+03,
1.60000000e+01],
[ 3.59384110e-01, 2.31281577e-01, 9.80787092e-01,
6.06335052e-01, 3.00000000e+00, 2.00000000e+00,
1.40000000e+01, 1.74614116e+02, 5.54365262e+02,
1.60000000e+01],
[ 7.01058769e-01, 8.78345353e-02, 5.30614298e-01,
8.06027846e-01, 3.00000000e+00, 0.00000000e+00,
1.00000000e+01, 2.77182631e+02, 4.93883301e+02,
1.60000000e+01],
[ 8.46154512e-01, 1.87416702e-01, 2.22742871e-01,
2.55321866e-01, 2.00000000e+00, 2.00000000e+00,
1.20000000e+01, 7.77817459e+01, 1.23470825e+02,
1.60000000e+01],
[ 2.44638450e-01, 4.89615141e-03, 1.44001771e-01,
3.93863621e-01, 5.00000000e+00, 0.00000000e+00,
1.20000000e+01, 4.93883301e+02, 5.87329536e+02,
1.60000000e+01],
[ 2.64003613e-01, 9.77735276e-01, 7.19114612e-01,
9.01769181e-01, 4.00000000e+00, 2.00000000e+00,
1.20000000e+01, 1.22498574e+01, 6.92956577e+01,
1.60000000e+01],
[ 9.76232131e-01, 4.73980819e-01, 8.56676586e-01,
1.85855418e-02, 4.00000000e+00, 3.00000000e+00,
1.10000000e+01, 3.08677063e+01, 6.22253967e+02,
1.60000000e+01],
[ 6.96363877e-01, 9.01892568e-01, 8.74152129e-02,
1.40171550e-01, 1.00000000e+00, 3.00000000e+00,
1.30000000e+01, 8.30609395e+02, 1.39691293e+03,
1.60000000e+01],
[ 2.44547789e-02, 6.95094086e-01, 6.19703209e-01,
4.80139384e-01, 1.00000000e+00, 3.00000000e+00,
1.40000000e+01, 1.83540480e+01, 5.19130872e+01,
1.60000000e+01],
[ 4.99682276e-01, 8.54642128e-02, 1.16964019e-01,
8.82417334e-01, 3.00000000e+00, 5.00000000e+00,
9.00000000e+00, 1.45676175e+01, 3.88908730e+01,
1.60000000e+01],
[ 3.32999431e-01, 7.58517863e-01, 9.57923895e-01,
8.58893091e-01, 4.00000000e+00, 5.00000000e+00,
1.20000000e+01, 2.77182631e+02, 5.23251131e+02,
1.60000000e+01],
[ 1.19481319e-01, 5.16393929e-01, 8.57334431e-03,
6.21834337e-01, 5.00000000e+00, 4.00000000e+00,
1.00000000e+01, 1.47997769e+03, 2.63702046e+03,
1.60000000e+01],
[ 7.87179804e-01, 9.29303701e-01, 5.02257370e-01,
9.89073623e-01, 2.00000000e+00, 4.00000000e+00,
1.30000000e+01, 8.80000000e+02, 1.31851023e+03,
1.60000000e+01],
[ 2.71883593e-01, 3.03523995e-01, 7.76082888e-01,
2.31191883e-01, 4.00000000e+00, 1.00000000e+00,
1.10000000e+01, 6.98456463e+02, 1.17465907e+03,
1.60000000e+01],
[ 3.25999117e-01, 3.93316786e-01, 5.96222219e-01,
1.59517751e-01, 0.00000000e+00, 3.00000000e+00,
9.00000000e+00, 1.09133822e+01, 1.64813778e+02,
1.60000000e+01],
[ 1.25434927e-01, 9.15497073e-01, 9.36152996e-01,
9.46849897e-01, 1.00000000e+00, 4.00000000e+00,
1.10000000e+01, 5.50000000e+01, 1.03826174e+02,
1.60000000e+01],
[ 2.20754135e-01, 2.08976731e-01, 1.98145901e-01,
8.55898171e-01, 4.00000000e+00, 4.00000000e+00,
1.20000000e+01, 4.93883301e+02, 7.39988845e+02,
1.60000000e+01],
[ 9.17049535e-01, 6.26543410e-03, 2.08145487e-02,
5.81651906e-01, 3.00000000e+00, 4.00000000e+00,
1.10000000e+01, 3.91995436e+02, 4.93883301e+02,
1.60000000e+01],
[ 3.44972430e-01, 6.08715328e-01, 9.36192542e-02,
9.67605131e-01, 5.00000000e+00, 2.00000000e+00,
1.00000000e+01, 2.61625565e+02, 2.93664768e+02,
1.60000000e+01],
[ 2.17571318e-01, 3.97113944e-01, 9.00572365e-01,
6.46500433e-01, 5.00000000e+00, 0.00000000e+00,
1.40000000e+01, 9.72271824e+00, 1.37500000e+01,
1.60000000e+01],
[ 6.09425226e-01, 9.53499121e-01, 3.67830864e-01,
2.95025276e-01, 1.00000000e+00, 4.00000000e+00,
1.10000000e+01, 1.09133822e+01, 3.72931009e+03,
1.60000000e+01],
[ 6.03153526e-01, 6.75552074e-01, 2.27855896e-01,
2.19524198e-01, 3.00000000e+00, 2.00000000e+00,
1.20000000e+01, 1.94454365e+01, 4.89994295e+01,
1.60000000e+01],
[ 4.44281858e-01, 5.36325911e-01, 7.90351741e-01,
7.57456099e-01, 0.00000000e+00, 3.00000000e+00,
1.30000000e+01, 5.50000000e+01, 1.64813778e+02,
1.60000000e+01],
[ 6.53895144e-01, 8.87003260e-01, 9.60499685e-03,
9.62404894e-01, 5.00000000e+00, 5.00000000e+00,
1.20000000e+01, 9.72271824e+00, 1.83540480e+01,
1.60000000e+01],
[ 9.17862960e-01, 7.84819867e-01, 6.00791650e-01,
4.39622459e-01, 0.00000000e+00, 2.00000000e+00,
1.00000000e+01, 4.40000000e+02, 6.22253967e+02,
1.60000000e+01],
[ 4.59416026e-01, 9.64323383e-01, 7.38905189e-03,
3.65515440e-01, 4.00000000e+00, 0.00000000e+00,
1.10000000e+01, 1.54338532e+01, 4.36535289e+01,
1.60000000e+01],
[ 2.18429455e-01, 8.03875937e-02, 6.71319582e-01,
7.06118679e-02, 0.00000000e+00, 2.00000000e+00,
1.30000000e+01, 6.59255114e+02, 1.24450793e+03,
1.60000000e+01],
[ 8.88572972e-01, 4.25046821e-01, 4.71270212e-01,
3.01855318e-01, 0.00000000e+00, 4.00000000e+00,
1.20000000e+01, 1.95997718e+02, 4.66163762e+02,
1.60000000e+01],
[ 7.06000732e-02, 5.00510665e-01, 2.49573626e-02,
2.57327761e-01, 1.00000000e+00, 1.00000000e+00,
1.10000000e+01, 3.11126984e+02, 1.04650226e+03,
1.60000000e+01],
[ 5.59480495e-01, 3.51992463e-01, 5.07058652e-01,
5.04548853e-01, 0.00000000e+00, 0.00000000e+00,
9.00000000e+00, 1.17465907e+03, 1.86465505e+03,
1.60000000e+01],
[ 3.04304972e-01, 7.25535293e-01, 7.81526949e-03,
9.95179416e-01, 1.00000000e+00, 4.00000000e+00,
1.10000000e+01, 1.09133822e+01, 1.03826174e+02,
1.60000000e+01],
[ 3.89979482e-01, 3.04110776e-01, 5.48185920e-01,
7.46558343e-01, 1.00000000e+00, 5.00000000e+00,
1.20000000e+01, 2.59565436e+01, 4.12034446e+01,
1.60000000e+01],
[ 6.03574550e-01, 7.71480041e-01, 6.40922688e-01,
5.57669701e-01, 3.00000000e+00, 0.00000000e+00,
1.00000000e+01, 2.18267645e+01, 3.46478289e+01,
1.60000000e+01],
[ 6.33824101e-01, 1.00791411e-01, 8.86926243e-01,
5.69302888e-01, 3.00000000e+00, 3.00000000e+00,
1.00000000e+01, 1.95997718e+02, 2.07652349e+02,
1.60000000e+01],
[ 4.83270844e-01, 4.31958188e-02, 8.89111208e-01,
3.03748410e-02, 5.00000000e+00, 1.00000000e+00,
1.10000000e+01, 6.17354127e+01, 8.73070579e+01,
1.60000000e+01],
[ 4.87945587e-01, 6.90624501e-01, 4.78157023e-01,
4.23797250e-02, 1.00000000e+00, 4.00000000e+00,
8.00000000e+00, 4.36535289e+01, 1.84997211e+02,
1.60000000e+01],
[ 4.17061098e-02, 8.54476414e-01, 2.04662729e-01,
2.41624398e-01, 1.00000000e+00, 3.00000000e+00,
9.00000000e+00, 1.10873052e+03, 2.09300452e+03,
1.60000000e+01],
[ 5.93181370e-01, 2.04257731e-01, 6.25243785e-01,
1.02961477e-01, 1.00000000e+00, 5.00000000e+00,
1.20000000e+01, 3.11126984e+02, 2.21746105e+03,
1.60000000e+01],
[ 7.75826478e-01, 2.84495709e-01, 8.59169508e-01,
6.31274988e-01, 2.00000000e+00, 2.00000000e+00,
1.10000000e+01, 9.17702400e+00, 1.09133822e+01,
1.60000000e+01],
[ 1.08605262e-01, 6.09662022e-01, 5.96985380e-01,
3.49630688e-01, 2.00000000e+00, 4.00000000e+00,
1.20000000e+01, 1.30812783e+02, 2.46941651e+02,
1.60000000e+01],
[ 1.20932073e-01, 6.81907879e-01, 9.21699988e-01,
1.67986030e-01, 3.00000000e+00, 0.00000000e+00,
1.20000000e+01, 8.17579892e+00, 4.18600904e+03,
1.60000000e+01],
[ 9.43480882e-01, 1.72716304e-01, 7.31615737e-01,
2.35318539e-01, 5.00000000e+00, 3.00000000e+00,
8.00000000e+00, 1.64813778e+02, 3.29627557e+02,
1.60000000e+01],
[ 3.36802048e-01, 8.93346266e-01, 5.56808351e-01,
4.23929893e-01, 4.00000000e+00, 0.00000000e+00,
8.00000000e+00, 3.88908730e+01, 3.95106641e+03,
1.60000000e+01],
[ 3.32561690e-01, 4.32786774e-01, 1.98115883e-01,
4.70694013e-03, 2.00000000e+00, 4.00000000e+00,
1.30000000e+01, 2.20000000e+02, 3.95106641e+03,
1.60000000e+01],
[ 7.74137628e-01, 3.24171250e-01, 9.54343651e-02,
6.80617300e-01, 5.00000000e+00, 5.00000000e+00,
1.20000000e+01, 1.10873052e+03, 3.95106641e+03,
1.60000000e+01],
[ 2.00390003e-01, 4.80073455e-01, 2.03514730e-01,
8.01010403e-01, 1.00000000e+00, 0.00000000e+00,
1.40000000e+01, 4.66163762e+02, 6.22253967e+02,
1.60000000e+01],
[ 3.88211952e-01, 7.23324070e-01, 9.08966653e-01,
1.15439206e-01, 5.00000000e+00, 2.00000000e+00,
1.10000000e+01, 6.54063913e+01, 8.73070579e+01,
1.60000000e+01],
[ 2.05755206e-02, 5.84655459e-01, 6.36086635e-01,
2.77855111e-01, 2.00000000e+00, 3.00000000e+00,
1.10000000e+01, 1.95997718e+02, 2.93664768e+02,
1.60000000e+01],
[ 4.92622980e-01, 9.89929425e-01, 9.76492480e-01,
8.92141956e-03, 2.00000000e+00, 1.00000000e+00,
1.40000000e+01, 5.87329536e+02, 1.86465505e+03,
1.60000000e+01],
[ 7.59947779e-01, 5.97140933e-02, 3.37852753e-02,
8.30424594e-01, 4.00000000e+00, 0.00000000e+00,
1.20000000e+01, 9.79988590e+01, 1.30812783e+02,
1.60000000e+01],
[ 2.33052433e-01, 3.95877031e-01, 7.80625965e-01,
6.12112506e-01, 0.00000000e+00, 4.00000000e+00,
1.20000000e+01, 3.88908730e+01, 1.23470825e+02,
1.60000000e+01],
[ 6.25775589e-01, 1.73320709e-01, 7.32627992e-01,
2.32759446e-01, 0.00000000e+00, 5.00000000e+00,
1.30000000e+01, 3.29627557e+02, 7.83990872e+02,
1.60000000e+01],
[ 6.13349483e-01, 4.70304177e-01, 2.07820805e-01,
1.58491173e-01, 2.00000000e+00, 1.00000000e+00,
1.40000000e+01, 2.07652349e+02, 2.46941651e+02,
1.60000000e+01],
[ 5.88138769e-01, 9.02429914e-01, 6.52229789e-01,
5.62310140e-01, 0.00000000e+00, 5.00000000e+00,
1.10000000e+01, 1.66121879e+03, 3.95106641e+03,
1.60000000e+01],
[ 2.32791422e-01, 2.67940554e-01, 8.00947476e-01,
2.51281146e-01, 4.00000000e+00, 3.00000000e+00,
1.30000000e+01, 3.88908730e+01, 7.77817459e+01,
1.60000000e+01],
[ 6.80249933e-02, 6.42926405e-02, 8.62884989e-02,
9.11257901e-01, 0.00000000e+00, 2.00000000e+00,
1.10000000e+01, 9.79988590e+01, 1.86465505e+03,
1.60000000e+01],
[ 4.63363314e-01, 7.89818197e-01, 6.07832977e-01,
1.10045903e-01, 3.00000000e+00, 3.00000000e+00,
1.20000000e+01, 8.66195722e+00, 1.37500000e+01,
1.60000000e+01],
[ 4.40565527e-02, 2.91823856e-01, 9.46863941e-02,
8.32175186e-01, 4.00000000e+00, 1.00000000e+00,
1.10000000e+01, 1.17465907e+03, 1.86465505e+03,
1.60000000e+01],
[ 4.59862069e-01, 3.49460539e-01, 9.07124286e-01,
8.39087962e-01, 5.00000000e+00, 4.00000000e+00,
1.30000000e+01, 8.17579892e+00, 6.54063913e+01,
1.60000000e+01],
[ 5.92218547e-01, 4.65846456e-01, 9.50627801e-01,
8.14859360e-01, 5.00000000e+00, 5.00000000e+00,
1.20000000e+01, 1.63515978e+01, 8.24068892e+01,
1.60000000e+01],
[ 3.01785372e-01, 9.53624952e-01, 6.82601271e-01,
1.26688128e-01, 2.00000000e+00, 5.00000000e+00,
1.10000000e+01, 6.17354127e+01, 5.23251131e+02,
1.60000000e+01],
[ 3.87658424e-01, 4.06595450e-01, 7.97436325e-01,
9.41025691e-01, 0.00000000e+00, 0.00000000e+00,
9.00000000e+00, 1.16540940e+02, 2.20000000e+02,
1.60000000e+01],
[ 8.17272867e-01, 7.40968764e-01, 8.04849419e-01,
6.91434430e-01, 3.00000000e+00, 2.00000000e+00,
1.00000000e+01, 3.46478289e+01, 1.23470825e+02,
1.60000000e+01]])
In [8]:
%time
for i, row in df.iterrows():
session = nonrealtimetools.Session()
builder = gk.generator.gendy1.make_builder(row)
out = gk.generator.gendy1.build_out(builder)
synthdef = builder.build()
with session.at(0):
synth_a = session.add_synth(duration=10, synthdef=synthdef)
gk.util.render_session(session, this_dir, row["hash"])
1 loop, best of 3: 18.9 s per loop
In [10]:
%timeit
for i, row in df.iterrows():
y, sr = librosa.load(os.path.join(this_dir, "aif_files", row["hash"] + ".aiff"))
_y_normed = librosa.util.normalize(y)
_mfcc = librosa.feature.mfcc(y=_y_normed, sr=sr, n_mfcc=13)
_cent = np.mean(librosa.feature.spectral_centroid(y=_y_normed, sr=sr))
_mfcc_mean = gk.feature_extraction.get_stats(_mfcc)["mean"]
X_row = np.append(_mfcc_mean, _cent)
if i==0:
X_mtx = X_row
else:
X_mtx = np.vstack((X_mtx, X_row))
1 loop, best of 3: 1min 28s per loop
Thought: For feature extraction, it would probably be faster to extract all time domain vectors $y$ into a NumPy array and perform the necessary LibROSA operations across the rows of the vector, possibly leveraging under-the-hood efficiencies.
"1min 43s per loop" below
In [13]:
for i, row in df.iterrows():
session = nonrealtimetools.Session()
builder = gk.generator.gendy1.make_builder(row)
out = gk.generator.gendy1.build_out(builder)
synthdef = builder.build()
with session.at(0):
synth_a = session.add_synth(duration=10, synthdef=synthdef)
gk.util.render_session(session, this_dir, row["hash"])
y, sr = librosa.load(os.path.join(this_dir, "aif_files", row["hash"] + ".aiff"))
_y_normed = librosa.util.normalize(y)
_mfcc = librosa.feature.mfcc(y=_y_normed, sr=sr, n_mfcc=13)
_cent = np.mean(librosa.feature.spectral_centroid(y=_y_normed, sr=sr))
_mfcc_mean = gk.feature_extraction.get_stats(_mfcc)["mean"]
X_row = np.append(_mfcc_mean, _cent)
if i==0:
X_mtx = X_row
else:
X_mtx = np.vstack((X_mtx, X_row))
1 loop, best of 3: 1min 43s per loop
In [16]:
X_mtx.shape
Out[16]:
(100, 14)
In [20]:
def col_rename_4_mfcc(c):
if (c < 13):
return "mfcc_mean_{}".format(c)
else:
return "spectral_centroid"
In [21]:
pd.DataFrame(X_mtx).rename_axis(lambda c: col_rename_4_mfcc(c), axis=1)
Out[21]:
mfcc_mean_0
mfcc_mean_1
mfcc_mean_2
mfcc_mean_3
mfcc_mean_4
mfcc_mean_5
mfcc_mean_6
mfcc_mean_7
mfcc_mean_8
mfcc_mean_9
mfcc_mean_10
mfcc_mean_11
mfcc_mean_12
spectral_centroid
0
-130.944498
212.246248
22.733654
40.690030
12.067444
11.575786
-2.923093
1.963526
1.012338
5.041066
-1.778145
-0.059208
-0.053824
439.129383
1
-399.895787
110.209896
74.756361
45.275353
30.022552
24.211951
21.498928
18.683785
15.939682
14.143797
13.070332
12.103905
11.177688
28.382782
2
91.596323
84.458611
-51.472184
22.995390
-13.212321
10.045572
-8.217880
5.379077
-5.687071
3.532893
-3.146553
7.163779
4.395984
2359.783056
3
108.614834
41.703153
-43.736769
12.669656
-12.150096
11.946690
5.544302
22.679438
6.981179
5.568756
-8.453263
5.247114
4.105417
3321.156070
4
-257.498754
196.193063
51.860213
33.223307
34.137235
20.907093
19.182403
15.203466
11.925751
11.138111
9.420953
8.718725
7.627488
185.699671
5
-193.675721
200.459160
9.867697
17.697053
-1.148573
13.137136
8.874220
2.058959
2.494816
7.649791
0.977434
3.701533
4.431168
281.375797
6
-99.255339
187.928813
-2.400399
14.618994
-13.758809
3.472774
0.320724
4.100003
-6.144052
6.141625
0.164230
1.029655
-0.117211
603.629813
7
-65.096548
119.754086
-35.833374
-13.031844
-13.445218
21.014780
-18.855273
-9.250680
12.133204
-11.209625
-19.034837
25.895658
-14.952459
1365.673027
8
-150.207907
199.412785
22.935211
46.877748
22.706417
26.449756
13.847491
7.392376
-9.558257
-17.755563
-19.497822
-4.273309
9.658858
432.687709
9
-149.157050
210.391977
22.183814
42.031267
18.932379
22.919038
11.152427
9.410636
-0.429824
-3.695098
-8.920190
-6.296199
-3.881610
399.406245
10
-340.201357
165.930251
70.565247
30.761530
31.261373
26.647257
18.388280
16.187891
14.585171
12.103353
11.079403
10.246418
9.120839
71.720036
11
6.112265
148.653894
-38.904657
15.262946
0.279633
1.558111
1.101243
3.030016
-1.815661
5.338824
-2.578960
5.812729
-2.170402
1217.617461
12
-6.867327
175.536196
-8.756827
-0.813062
-25.938709
13.655398
8.412102
-5.728035
-15.097791
17.403257
-3.629439
-7.787881
3.420220
1074.711978
13
-273.189933
196.842472
44.006941
32.859688
30.937577
15.734702
11.308977
1.564761
-1.615650
1.674830
5.060813
5.387484
1.083436
138.239326
14
-1.179280
186.504575
-13.168122
7.702715
-6.959950
12.097020
-9.296268
0.670239
-1.656644
-1.433505
-4.144156
2.415664
-5.776426
966.652696
15
4.551287
136.437145
-51.754995
-7.027945
-4.988497
10.530632
-12.529925
6.626493
-3.929798
1.137125
-2.796705
3.836182
-2.504655
1328.003713
16
85.181769
119.017278
-55.916597
12.750346
-6.694964
1.678898
-2.779380
0.952695
-2.999644
1.798908
-4.016067
2.047174
-4.190821
1928.751797
17
90.191236
14.656168
-47.594906
5.551048
-17.733038
0.384498
-8.836457
11.887268
13.141246
27.478729
11.806024
5.389236
-10.406571
3692.915963
18
-83.394020
187.360281
20.232888
38.943839
-4.743480
-10.343549
-22.295417
6.048916
14.443237
7.609337
-19.994906
-6.595278
8.422725
716.223272
19
-73.318825
171.378497
-22.353387
-1.984178
-8.114975
23.694265
-4.649516
-0.134339
8.430598
6.442983
-3.987355
12.046813
-1.915677
777.648864
20
18.196390
36.265611
-57.300253
34.892149
-22.674767
7.679253
-10.483903
-2.425615
-0.539621
13.252110
31.706520
32.270856
15.487768
2876.427698
21
-202.216136
209.230301
34.720954
38.528029
26.253699
19.229712
17.196148
11.705229
11.479854
8.205794
6.759941
3.380408
0.222500
282.359030
22
-313.176685
155.511469
73.811454
32.795860
29.677764
27.192494
19.588297
15.911409
14.503815
12.399183
10.834480
9.971129
9.036276
108.303910
23
-9.615172
129.716910
-52.212949
-1.136605
13.376242
-4.356900
-18.290028
29.557186
-34.970057
26.575242
-15.458440
0.662723
-0.393250
1729.641814
24
-259.514615
185.040826
59.646164
30.595194
34.250743
22.800958
18.093277
16.527335
12.549139
11.291874
10.078129
8.796024
8.328326
172.221578
25
-35.086315
136.811693
-32.048640
-8.674571
-17.973700
17.048128
-7.139772
-12.386438
-0.602572
7.252857
-16.222938
8.550010
-0.207872
1251.268593
26
2.140036
22.958742
-61.243439
31.232914
-18.617048
-1.252628
-2.374840
-8.556711
-5.999625
-3.039336
15.848294
32.565206
34.893593
2883.351422
27
-180.303722
210.031939
25.614925
42.866791
21.707217
23.702508
15.505624
14.207790
8.508061
4.411809
-1.929581
-6.125128
-8.086553
320.154860
28
109.403837
69.055065
-16.386115
14.482498
-7.015294
7.962498
-3.539384
7.110676
-3.415325
4.134727
-3.697713
4.036978
-3.090759
3040.898005
29
-105.319927
188.452663
-9.875766
9.693579
-1.902148
16.048841
-4.273755
3.942981
3.458656
2.168673
1.654555
5.832600
0.048751
563.021922
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
70
-307.813300
171.501835
43.073483
1.186148
-17.059571
-4.280283
26.887092
15.969707
-15.146229
-1.326479
25.602239
4.725988
-7.229602
102.412996
71
-256.262220
177.667570
35.279755
41.074118
32.646052
22.869990
7.365998
-2.746482
-11.832048
-6.774149
4.203519
12.727086
8.279531
194.582872
72
-123.638507
196.101218
16.091192
37.703966
8.586197
6.572689
-8.640426
-0.573421
2.027123
6.996951
-3.244056
-2.842037
-1.415446
547.827122
73
91.201402
61.890904
-50.296706
34.611358
-19.843078
34.275911
5.463908
26.850593
-12.961399
-1.970077
-4.427944
23.936790
2.149188
3021.571060
74
61.521574
75.549878
-70.206479
4.072889
-12.268407
-0.804737
-8.762767
1.657496
-10.471751
-2.039512
-8.164378
1.220309
-0.395788
2299.984637
75
-348.854984
131.896991
77.298459
38.691859
27.370018
25.832986
21.517795
16.635897
14.360958
13.147170
11.757687
10.624068
9.748905
69.675077
76
-3.951955
175.926868
-16.887992
2.177106
-21.221609
8.061110
1.376236
-0.831991
-8.723344
6.918125
-6.161587
2.025728
-1.955819
1068.693765
77
6.312709
64.119320
0.662811
20.428846
2.098026
9.363664
-1.253574
7.691288
0.152815
6.425648
-0.281768
6.772991
0.180943
2403.080959
78
-265.967826
189.684948
40.921330
28.918336
3.856197
-16.878804
-8.641422
11.581525
22.577539
1.036239
-14.762041
3.065197
17.231147
160.792166
79
114.066912
65.705402
-13.982049
15.738120
-2.697014
12.161346
-0.194070
7.555816
-1.360006
6.779690
-1.487610
5.401036
-1.348300
3137.934756
80
52.080801
-2.029646
-9.214531
16.661814
14.812910
28.503113
7.405197
0.958969
-1.424717
17.050482
5.375132
2.593108
0.867433
4607.674790
81
0.129542
1.296577
1.134177
29.180047
16.657870
13.876821
-5.846684
14.828635
13.349497
2.471706
-3.025962
17.216665
0.068455
3728.938367
82
50.185707
69.750195
-62.128786
31.666168
-12.838869
-0.678508
3.423202
-9.808444
2.928664
-8.830149
4.075028
7.549014
20.555221
2545.937078
83
-141.369083
200.989763
22.828728
47.725317
23.799062
24.674824
5.827063
-2.230552
-16.662661
-13.424836
-4.548865
10.181220
9.344715
427.742851
84
-33.911887
177.238500
1.460917
-1.179384
-29.288186
9.006010
16.565548
-4.567549
-23.957549
18.389237
7.319709
-18.177522
3.227128
1033.168160
85
95.620136
24.162909
-21.330590
12.638309
-5.323904
19.195329
12.701693
19.753890
-3.804634
-5.664643
-4.645723
17.033040
6.003195
4041.866233
86
-78.124385
180.139992
1.803036
26.684432
-10.907147
-10.352879
-19.287853
2.790099
3.456098
-2.008889
-21.425404
-8.388509
0.576494
809.815892
87
-102.630029
208.396622
18.649455
38.064037
7.297848
6.976391
-8.479380
-0.673444
-0.535007
4.607480
-4.505434
-2.000420
-1.579127
536.744984
88
36.237315
81.456456
-68.728021
6.888781
0.927624
-3.934688
-7.373528
6.245210
-12.398505
7.953072
-8.427711
1.291764
-8.398082
2154.611718
89
-13.889859
152.856319
-21.261154
-13.211758
-10.707592
27.258129
-17.125446
-8.789949
14.124429
-4.738955
-14.482688
19.706406
-16.619446
1301.327022
90
83.447113
1.812704
4.520221
28.265909
15.011482
6.466598
-8.997236
16.023643
7.023527
-6.226134
-0.380711
14.075010
-10.386671
4891.810972
91
-185.604592
206.337093
25.511371
45.913118
24.236532
25.398985
16.804346
16.984858
12.793724
9.822169
1.918410
-7.251096
-15.291040
334.386811
92
106.618408
68.227289
-35.364793
11.350113
-13.695992
3.597132
-9.584854
3.928110
-4.155313
6.291189
-1.728305
5.843339
-1.471005
2930.730071
93
-413.286972
103.664304
74.358332
46.970740
31.255063
25.115325
22.120139
18.952836
15.974466
13.969859
12.764401
11.691259
10.617216
27.406534
94
45.976830
-14.512710
-13.656387
18.141022
8.774175
36.462065
11.854833
0.298979
-15.362854
20.456249
22.244089
1.058561
-17.340131
4586.614343
95
-219.990763
204.620151
41.814556
36.895815
30.497998
20.563205
18.753211
11.094620
6.271293
1.090289
-0.426076
1.362617
3.783278
212.787669
96
-224.056647
211.048163
36.244777
40.204388
27.123646
20.813520
16.674609
10.407743
7.226611
3.281310
2.596085
1.825429
1.610101
212.826591
97
-61.565860
204.165510
8.677969
26.727983
-1.837981
8.164580
-4.215543
1.924936
-4.988670
2.121297
-3.525907
1.950059
-2.732152
660.369735
98
-50.747979
192.216719
1.366500
20.313345
-14.808932
-7.260769
-10.273262
7.347047
-2.840807
-3.044787
-5.123745
3.984245
-4.706193
796.513395
99
-236.292400
206.459818
34.469195
37.975486
24.656601
13.375484
6.655611
-1.206207
1.702022
4.473805
5.835302
1.604360
-0.253468
195.427304
100 rows × 14 columns
In [23]:
from sklearn import linear_model
from sklearn import model_selection
from sklearn import preprocessing
In [24]:
import sklearn as sk
In [39]:
import matplotlib.pyplot as plt
%matplotlib inline
In [66]:
pmtx.shape
Out[66]:
(100, 10)
In [67]:
X_mtx.shape
Out[67]:
(100, 14)
In [68]:
X_mtx[0]
Out[68]:
array([ -1.30944498e+02, 2.12246248e+02, 2.27336540e+01,
4.06900302e+01, 1.20674442e+01, 1.15757863e+01,
-2.92309292e+00, 1.96352599e+00, 1.01233809e+00,
5.04106581e+00, -1.77814489e+00, -5.92080878e-02,
-5.38241320e-02, 4.39129383e+02])
In [31]:
X_train, X_test, y_train, y_test = sk.model_selection.train_test_split(
X_mtx, pmtx, test_size=0.4, random_state=1)
In [38]:
# Create linear regression objectc
regr = linear_model.LinearRegression()
# Train the model using the training sets
regr.fit(X_train, y_train)
# The coefficients
print('Coefficients: \n', regr.coef_)
# The mean squared error
print("Mean squared error: %.2f"
% np.mean((regr.predict(X_test) - y_test) ** 2))
# Explained variance score: 1 is perfect prediction
print('Variance score: %.2f' % regr.score(X_test, y_test))
Coefficients:
[[ -7.64326344e-04 5.43840972e-04 8.33835578e-04 -1.63296951e-03
1.84396630e-03 -1.63494109e-03 -4.13973644e-03 9.93769874e-03
-1.21937858e-02 1.07145282e-02 -3.50315085e-03 8.43229154e-03
5.97796792e-03 7.19034767e-05]
[ 4.14076160e-03 1.49650267e-03 7.37848988e-03 -5.89920432e-03
6.43978853e-03 2.56377513e-03 5.08002799e-03 6.34448830e-03
2.48995221e-03 -9.09774452e-04 3.85030377e-03 1.95772826e-03
1.02450172e-02 -8.23419500e-05]
[ -1.58993713e-03 2.26292332e-03 -2.75211523e-03 7.45264689e-03
-9.26048513e-04 -1.65945356e-03 -4.77375839e-03 1.51891806e-02
-1.34561050e-02 1.02081968e-02 -5.44665508e-03 1.38347150e-02
-1.23215340e-02 1.37929495e-04]
[ 7.85798120e-04 2.41554881e-03 1.78210384e-03 -9.60912755e-04
-5.29459040e-03 3.47813067e-03 -1.96233126e-03 2.84434039e-03
1.35618930e-03 -1.21362974e-03 2.00095401e-03 1.95225907e-03
7.85679139e-03 4.24680697e-05]
[ 7.59350445e-03 -1.94676177e-02 4.10676580e-05 -1.23607006e-02
8.17922958e-03 1.16830267e-03 2.88827243e-02 -5.40019464e-02
5.85324462e-02 -3.58495151e-02 2.00269831e-02 -5.54215649e-02
3.18534999e-02 -1.65913946e-03]
[ -7.57996905e-04 1.48878162e-02 -2.02212716e-02 -1.26859790e-02
3.28634748e-02 -1.00350966e-02 2.60161747e-02 3.03667595e-02
1.97929204e-02 -5.33650786e-03 1.85626625e-02 1.78251979e-04
-1.00894504e-02 3.90389290e-04]
[ -7.42013192e-03 1.64585909e-02 -3.66278515e-02 -1.15205242e-02
6.69830320e-03 2.32797054e-02 -1.70235246e-02 4.02606929e-02
5.74493635e-02 -5.28742501e-02 1.69008951e-02 2.07720815e-02
1.42142293e-02 6.60588879e-04]
[ -4.02348887e+00 4.78619594e+00 -5.20732849e+00 1.43378482e+00
-1.86539229e-01 -8.95742471e-01 -2.38130301e+00 -1.14239048e-01
-7.14676019e+00 6.04096194e+00 -5.28217010e+00 6.65615280e+00
-4.69621834e+00 6.35692181e-01]
[ 1.10679242e+01 -1.45142504e+01 2.78669011e+01 -1.57162603e+01
2.37260432e+01 -2.25193655e+01 1.09819093e+01 -2.93601263e+01
1.87695403e+01 -2.66324515e+01 2.66129494e+01 -3.35883337e+01
-9.61925708e+00 -2.26352472e-01]
[ 0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00
0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00
0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00
0.00000000e+00 0.00000000e+00]]
Mean squared error: 80571.86
Variance score: 0.27
/Users/spacecoffin/Development/sprbrg/lib/python3.5/site-packages/scipy/linalg/basic.py:884: RuntimeWarning: internal gelsd driver lwork query error, required iwork dimension not returned. This is likely the result of LAPACK bug 0038, fixed in LAPACK 3.2.2 (released July 21, 2010). Falling back to 'gelss' driver.
warnings.warn(mesg, RuntimeWarning)
In [42]:
# Scale data
standard_scaler = sk.preprocessing.StandardScaler()
X_scaled = standard_scaler.fit_transform(X_mtx)
#Xte_s = standard_scaler.transform(X_test)
robust_scaler = sk.preprocessing.RobustScaler()
X_rscaled = robust_scaler.fit_transform(X_mtx)
#Xte_r = robust_scaler.transform(X_test)
In [57]:
X_scaled.mean(axis=0)
Out[57]:
array([ 2.95319325e-16, 1.93178806e-16, 3.21964677e-17,
-5.66213743e-17, -6.66133815e-18, 3.96696564e-16,
5.88418203e-17, -1.86517468e-16, 7.00828284e-17,
3.03090886e-16, -1.66533454e-17, 7.10542736e-17,
-1.66533454e-16, 4.86277685e-16])
In [59]:
X_scaled.mean(axis=0).mean()
Out[59]:
1.0526797687506115e-16
In [61]:
X_scaled.std(axis=0)
Out[61]:
array([ 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.])
In [43]:
X_train, X_test, y_train, y_test = sk.model_selection.train_test_split(
X_scaled, pmtx, test_size=0.4, random_state=1)
In [44]:
# Create linear regression objectc
regr = linear_model.LinearRegression()
# Train the model using the training sets
regr.fit(X_train, y_train)
# The coefficients
print('Coefficients: \n', regr.coef_)
# The mean squared error
print("Mean squared error: %.2f"
% np.mean((regr.predict(X_test) - y_test) ** 2))
# Explained variance score: 1 is perfect prediction
print('Variance score: %.2f' % regr.score(X_test, y_test))
Coefficients:
[[ -1.16321270e-01 3.82759545e-02 3.54522472e-02 -2.71379638e-02
3.49357740e-02 -1.90305103e-02 -5.18060153e-02 1.02756139e-01
-1.37969854e-01 1.01095033e-01 -3.91266879e-02 7.60528902e-02
5.51004476e-02 1.01775935e-01]
[ 6.30174076e-01 1.05325032e-01 3.13711785e-01 -9.80375889e-02
1.22008193e-01 2.98420226e-02 6.35731312e-02 6.56022225e-02
2.81732309e-02 -8.58401572e-03 4.30040385e-02 1.76572278e-02
9.44309239e-02 -1.16551095e-01]
[ -2.41969294e-01 1.59266319e-01 -1.17011881e-01 1.23853912e-01
-1.75449093e-02 -1.93158324e-02 -5.97403733e-02 1.57056638e-01
-1.52252703e-01 9.63176327e-02 -6.08336845e-02 1.24778663e-01
-1.13570706e-01 1.95232608e-01]
[ 1.19589016e-01 1.70008221e-01 7.57698369e-02 -1.59691993e-02
-1.00311277e-01 4.04850070e-02 -2.45572550e-02 2.94105753e-02
1.53449669e-02 -1.14509885e-02 2.23486532e-02 1.76078998e-02
7.24180405e-02 6.01115230e-02]
[ 1.15563998e+00 -1.37014621e+00 1.74607657e-03 -2.05419785e-01
1.54963632e-01 1.35988973e-02 3.61447856e-01 -5.58381942e-01
6.62281034e-01 -3.38251751e-01 2.23681353e-01 -4.99860587e-01
2.93601793e-01 -2.34843261e+00]
[ -1.15358005e-01 1.04781619e+00 -8.59749258e-01 -2.10825516e-01
6.22631187e-01 -1.16807272e-01 3.25574916e-01 3.13993315e-01
2.23952297e-01 -5.03516749e-02 2.07326357e-01 1.60769800e-03
-9.29970255e-02 5.52577382e-01]
[ -1.12925477e+00 1.15836854e+00 -1.55730900e+00 -1.91457077e-01
1.26906010e-01 2.70972866e-01 -2.13037953e-01 4.16296919e-01
6.50026205e-01 -4.98885623e-01 1.88766079e-01 1.87348461e-01
1.31016160e-01 9.35031986e-01]
[ -6.12326578e+02 3.36856227e+02 -2.21400360e+02 2.38277569e+01
-3.53417106e+00 -1.04263306e+01 -2.98004044e+01 -1.18123560e+00
-8.08639317e+01 5.69984265e+01 -5.89965520e+01 6.00334626e+01
-4.32862365e+01 8.99791901e+02]
[ 1.68440485e+03 -1.02152433e+03 1.18481904e+03 -2.61185099e+02
4.49513465e+02 -2.62122605e+02 1.37431204e+02 -3.03584694e+02
2.12372989e+02 -2.51285780e+02 2.97240003e+02 -3.02941359e+02
-8.86631341e+01 -3.20391106e+02]
[ 0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00
0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00
0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00
0.00000000e+00 0.00000000e+00]]
Mean squared error: 80571.86
Variance score: 0.27
In [64]:
y_test[0]
Out[64]:
array([ 3.32561690e-01, 4.32786774e-01, 1.98115883e-01,
4.70694013e-03, 2.00000000e+00, 4.00000000e+00,
1.30000000e+01, 2.20000000e+02, 3.95106641e+03,
1.60000000e+01])
In [71]:
X_test[0]
Out[71]:
array([ 1.01086499, -1.90257535, -0.26937053, -0.36982032, 0.42170525,
1.26344082, 0.32565618, -0.63435796, -0.34336578, 1.15169182,
0.46333517, -0.35546419, -0.22943205, 2.27780905])
In [70]:
regr.predict(X_test[0])
/Users/spacecoffin/Development/sprbrg/lib/python3.5/site-packages/sklearn/utils/validation.py:395: DeprecationWarning: Passing 1d arrays as data is deprecated in 0.17 and will raise ValueError in 0.19. Reshape your data either using X.reshape(-1, 1) if your data has a single feature or X.reshape(1, -1) if it contains a single sample.
DeprecationWarning)
Out[70]:
array([[ 5.81091297e-01, 6.46632233e-01, 4.05079195e-01,
3.24762723e-01, 1.28734484e+00, 1.97150205e+00,
1.01313047e+01, 1.11999703e+03, 3.38978192e+03,
1.60000000e+01]])
In [72]:
y_test[0]
Out[72]:
array([ 3.32561690e-01, 4.32786774e-01, 1.98115883e-01,
4.70694013e-03, 2.00000000e+00, 4.00000000e+00,
1.30000000e+01, 2.20000000e+02, 3.95106641e+03,
1.60000000e+01])
In [ ]:
Content source: spacecoffin/GravelKicker
Similar notebooks: