deepof_model_evaluation.ipynb 1.46 MB
Newer Older
1
2
3
4
{
 "cells": [
  {
   "cell_type": "code",
5
   "execution_count": 1,
6
   "metadata": {},
7
   "outputs": [],
8
9
10
11
12
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
13
14
  {
   "cell_type": "code",
15
   "execution_count": 2,
16
17
18
19
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
20
    "\n",
21
22
23
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# deepOF model evaluation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Given a dataset and a trained model, this notebook allows the user to \n",
    "\n",
    "* Load and inspect the different models (encoder, decoder, grouper, gmvaep)\n",
    "* Visualize reconstruction quality for a given model\n",
    "* Visualize a static latent space\n",
    "* Visualize trajectories on the latent space for a given video\n",
    "* sample from the latent space distributions and generate video clips showcasing generated data"
   ]
  },
  {
   "cell_type": "code",
46
   "execution_count": 3,
47
48
49
50
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
51
    "\n",
52
53
54
55
56
    "os.chdir(os.path.dirname(\"../\"))"
   ]
  },
  {
   "cell_type": "code",
lucas_miranda's avatar
lucas_miranda committed
57
   "execution_count": 4,
58
59
60
61
62
63
64
   "metadata": {},
   "outputs": [],
   "source": [
    "import deepof.data\n",
    "import deepof.utils\n",
    "import numpy as np\n",
    "import pandas as pd\n",
65
    "import re\n",
66
    "import tensorflow as tf\n",
67
    "from collections import Counter\n",
68
69
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
70
71
72
73
74
    "from sklearn.manifold import TSNE\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
    "import umap\n",
    "\n",
75
    "from ipywidgets import interactive, interact, HBox, Layout, VBox\n",
76
77
    "from IPython import display\n",
    "from matplotlib.animation import FuncAnimation\n",
lucas_miranda's avatar
lucas_miranda committed
78
    "from mpl_toolkits.mplot3d import Axes3D\n",
79
80
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
lucas_miranda's avatar
lucas_miranda committed
81
    "import tqdm.notebook as tqdm\n",
82
83
84
85
86
87
88
89
90
91
92
93
94
    "\n",
    "from ipywidgets import interact"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. Define and run project"
   ]
  },
  {
   "cell_type": "code",
lucas_miranda's avatar
lucas_miranda committed
95
   "execution_count": 5,
96
97
98
   "metadata": {},
   "outputs": [],
   "source": [
99
    "path = os.path.join(\"..\", \"..\", \"Desktop\", \"deepof-data\", \"deepof_single_topview\")\n",
100
    "trained_network = os.path.join(\"..\", \"..\", \"Desktop\", \"deepof_trained_weights_280521\", \"var_annealing\")\n",
101
    "exclude_bodyparts = tuple([\"\"])\n",
102
    "window_size = 22\n",
103
    "batch_size = 32"
104
105
106
107
   ]
  },
  {
   "cell_type": "code",
lucas_miranda's avatar
lucas_miranda committed
108
   "execution_count": 6,
109
   "metadata": {},
110
111
112
113
114
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
lucas_miranda's avatar
lucas_miranda committed
115
116
      "CPU times: user 44.8 s, sys: 3.16 s, total: 48 s\n",
      "Wall time: 40.2 s\n"
117
118
119
     ]
    }
   ],
120
121
122
   "source": [
    "%%time\n",
    "proj = deepof.data.project(\n",
123
    "    path=path, smooth_alpha=0.999, exclude_bodyparts=exclude_bodyparts, arena_dims=[380],\n",
124
125
126
127
128
    ")"
   ]
  },
  {
   "cell_type": "code",
lucas_miranda's avatar
lucas_miranda committed
129
   "execution_count": 7,
130
131
132
   "metadata": {
    "scrolled": true
   },
133
134
135
136
137
138
139
140
141
142
143
144
145
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading trajectories...\n",
      "Smoothing trajectories...\n",
      "Interpolating outliers...\n",
      "Iterative imputation of ocluded bodyparts...\n",
      "Computing distances...\n",
      "Computing angles...\n",
      "Done!\n",
      "deepof analysis of 166 videos\n",
lucas_miranda's avatar
lucas_miranda committed
146
147
      "CPU times: user 11min 35s, sys: 18.4 s, total: 11min 54s\n",
      "Wall time: 2min 32s\n"
148
149
150
     ]
    }
   ],
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
   "source": [
    "%%time\n",
    "proj = proj.run(verbose=True)\n",
    "print(proj)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Load pretrained deepof model"
   ]
  },
  {
   "cell_type": "code",
lucas_miranda's avatar
lucas_miranda committed
166
   "execution_count": 8,
167
168
169
   "metadata": {},
   "outputs": [],
   "source": [
170
    "coords = proj.get_coords(center=\"Center\", align=\"Spine_1\", align_inplace=True)\n",
171
    "data_prep = coords.preprocess(test_videos=0, window_step=1, window_size=window_size, shuffle=False)[\n",
172
173
    "    0\n",
    "]"
174
175
176
177
   ]
  },
  {
   "cell_type": "code",
lucas_miranda's avatar
lucas_miranda committed
178
   "execution_count": 9,
179
   "metadata": {},
180
181
182
183
   "outputs": [
    {
     "data": {
      "text/plain": [
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
       "['GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=7_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=1_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=3_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=5_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=2_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=8_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=10_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=5_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=20_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=15_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=9_final_weights.h5',\n",
       " 'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.15_PhenoPred=0.0_RuleBasedPred=0.15_loss=ELBO_loss_warmup=10_warmup_mode=sigmoid_encoding=6_k=15_latreg=variance_entknn=100_run=4_final_weights.h5']"
584
585
      ]
     },
lucas_miranda's avatar
lucas_miranda committed
586
     "execution_count": 9,
587
588
589
590
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
591
592
593
594
595
596
   "source": [
    "[i for i in os.listdir(trained_network) if i.endswith(\"h5\")]"
   ]
  },
  {
   "cell_type": "code",
lucas_miranda's avatar
lucas_miranda committed
597
   "execution_count": 10,
598
   "metadata": {},
599
600
601
602
   "outputs": [
    {
     "data": {
      "text/plain": [
603
       "'GMVAE_input_type=coords_window_size=22_NextSeqPred=0.0_PhenoPred=0.0_RuleBasedPred=0.0_loss=ELBO_loss_warmup=25_warmup_mode=linear_encoding=6_k=15_latreg=variance_entknn=100_run=6_final_weights.h5'"
604
605
      ]
     },
lucas_miranda's avatar
lucas_miranda committed
606
     "execution_count": 10,
607
608
609
610
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
611
   "source": [
612
    "deepof_weights = [i for i in os.listdir(trained_network) if i.endswith(\"h5\")][1]\n",
613
614
615
616
617
    "deepof_weights"
   ]
  },
  {
   "cell_type": "code",
lucas_miranda's avatar
lucas_miranda committed
618
   "execution_count": 11,
619
620
621
   "metadata": {},
   "outputs": [],
   "source": [
622
623
624
625
    "# Set model parameters\n",
    "encoding = int(re.findall(\"encoding=(\\d+)_\", deepof_weights)[0])\n",
    "k = int(re.findall(\"k=(\\d+)_\", deepof_weights)[0])\n",
    "loss = re.findall(\"loss=(.+?)_\", deepof_weights)[0]\n",
626
627
628
    "NextSeqPred = float(re.findall(\"NextSeqPred=(.+?)_\", deepof_weights)[0])\n",
    "PhenoPred = float(re.findall(\"PhenoPred=(.+?)_\", deepof_weights)[0])\n",
    "RuleBasedPred = float(re.findall(\"RuleBasedPred=(.+?)_\", deepof_weights)[0])"
629
630
631
632
   ]
  },
  {
   "cell_type": "code",
lucas_miranda's avatar
lucas_miranda committed
633
   "execution_count": 12,
634
   "metadata": {},
635
   "outputs": [],
636
   "source": [
637
    "(\n",
638
    "    encoder,\n",
639
640
641
642
643
    "    decoder,\n",
    "    grouper,\n",
    "    gmvaep,\n",
    "    prior,\n",
    "    posterior,\n",
644
    ") = deepof.models.GMVAE(\n",
645
646
647
    "    loss=loss,\n",
    "    number_of_components=k,\n",
    "    compile_model=True,\n",
648
    "    batch_size=batch_size,\n",
649
    "    encoding=encoding,\n",
650
651
652
    "    next_sequence_prediction=NextSeqPred,\n",
    "    phenotype_prediction=PhenoPred,\n",
    "    rule_based_prediction=RuleBasedPred,\n",
653
654
655
    ").build(\n",
    "    data_prep.shape\n",
    ")\n",
656
    "gmvaep.load_weights(os.path.join(trained_network, deepof_weights))"
657
658
   ]
  },
lucas_miranda's avatar
lucas_miranda committed
659
660
  {
   "cell_type": "code",
lucas_miranda's avatar
lucas_miranda committed
661
   "execution_count": 14,
lucas_miranda's avatar
lucas_miranda committed
662
663
664
   "metadata": {},
   "outputs": [
    {
lucas_miranda's avatar
lucas_miranda committed
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
     "data": {
      "text/plain": [
       "array([[[-0.02210353, -0.785747  , -1.008676  , ...,  0.6001921 ,\n",
       "          0.04166124,  0.2981286 ],\n",
       "        [-0.02017598, -0.7876975 , -1.0585053 , ...,  0.5523039 ,\n",
       "          0.02210494,  0.23301357],\n",
       "        [-0.02075703, -0.82487637, -1.1057181 , ...,  0.57362604,\n",
       "         -0.02356116,  0.22122964],\n",
       "        ...,\n",
       "        [-0.00629616, -0.82951194, -1.1901224 , ...,  0.63662106,\n",
       "         -0.33436167,  0.3424397 ],\n",
       "        [-0.00440669, -0.8475035 , -1.1641558 , ...,  0.67850864,\n",
       "         -0.34476152,  0.3831051 ],\n",
       "        [-0.00504745, -0.88264775, -1.1671872 , ...,  0.7463107 ,\n",
       "         -0.36547643,  0.4086566 ]],\n",
       "\n",
       "       [[-0.0227385 , -0.7558565 , -1.0477355 , ...,  0.59044003,\n",
       "         -0.14209464,  0.30623406],\n",
       "        [-0.02094234, -0.7513252 , -1.1023679 , ...,  0.54719466,\n",
       "         -0.16325517,  0.2565396 ],\n",
       "        [-0.02090328, -0.7864068 , -1.1496762 , ...,  0.5761403 ,\n",
       "         -0.19885018,  0.2585564 ],\n",
       "        ...,\n",
       "        [-0.00962175, -0.7755053 , -1.2331941 , ...,  0.57873404,\n",
       "         -0.46255097,  0.35361302],\n",
       "        [-0.00706752, -0.7978875 , -1.2086567 , ...,  0.62388206,\n",
       "         -0.4573907 ,  0.39810222],\n",
       "        [-0.00935922, -0.8508443 , -1.2035465 , ...,  0.6940046 ,\n",
       "         -0.47271004,  0.4187271 ]],\n",
       "\n",
       "       [[-0.02460377, -0.71985835, -1.0561382 , ...,  0.57191074,\n",
       "         -0.17259115,  0.3035625 ],\n",
       "        [-0.02447619, -0.71244407, -1.1156678 , ...,  0.5265259 ,\n",
       "         -0.19369215,  0.2462604 ],\n",
       "        [-0.02500374, -0.74148995, -1.1665585 , ...,  0.55580366,\n",
       "         -0.22961235,  0.256171  ],\n",
       "        ...,\n",
       "        [-0.01054991, -0.7446184 , -1.0810308 , ...,  0.5598789 ,\n",
       "         -0.2368857 ,  0.29076478],\n",
       "        [-0.00638657, -0.7622384 , -1.059313  , ...,  0.5997677 ,\n",
       "         -0.22255218,  0.3153192 ],\n",
       "        [-0.00680301, -0.8162386 , -1.0573621 , ...,  0.66828907,\n",
       "         -0.21515116,  0.3255418 ]],\n",
       "\n",
       "       ...,\n",
       "\n",
       "       [[-0.02382827, -0.81191695, -1.0895592 , ...,  0.6342349 ,\n",
       "         -0.64975667,  0.43953937],\n",
       "        [-0.02668396, -0.81807435, -1.179942  , ...,  0.6350413 ,\n",
       "         -0.72981477,  0.46332645],\n",
       "        [-0.03092471, -0.834882  , -1.2502159 , ...,  0.64505184,\n",
       "         -0.79759383,  0.537423  ],\n",
       "        ...,\n",
       "        [ 0.01370875, -0.67243356, -0.2821848 , ...,  0.41080245,\n",
       "          0.5612177 ,  0.2569335 ],\n",
       "        [ 0.01754688, -0.6227588 , -0.27820545, ...,  0.3890058 ,\n",
       "          0.67694294,  0.264346  ],\n",
       "        [ 0.01354074, -0.6513146 , -0.2939797 , ...,  0.46852356,\n",
       "          0.7120316 ,  0.27343553]],\n",
       "\n",
       "       [[-0.02230327, -0.8834006 , -1.0685817 , ...,  0.67149925,\n",
       "         -0.63458157,  0.4445827 ],\n",
       "        [-0.02467555, -0.9010228 , -1.1511382 , ...,  0.6756688 ,\n",
       "         -0.71627903,  0.46900186],\n",
       "        [-0.02870812, -0.9344835 , -1.2167535 , ...,  0.6911353 ,\n",
       "         -0.782382  ,  0.54373497],\n",
       "        ...,\n",
       "        [ 0.02096284, -0.6742392 , -0.15289709, ...,  0.42350322,\n",
       "          0.7991308 ,  0.33673635],\n",
       "        [ 0.0233632 , -0.61396956, -0.1666486 , ...,  0.39223164,\n",
       "          0.8993046 ,  0.35189033],\n",
       "        [ 0.01740324, -0.6450024 , -0.19953743, ...,  0.4831066 ,\n",
       "          0.8951427 ,  0.3776582 ]],\n",
       "\n",
       "       [[-0.02463406, -0.8174598 , -1.0923108 , ...,  0.631325  ,\n",
       "         -0.5889373 ,  0.4615081 ],\n",
       "        [-0.02725769, -0.8099975 , -1.1758723 , ...,  0.6114285 ,\n",
       "         -0.660674  ,  0.4733273 ],\n",
       "        [-0.03179519, -0.81105834, -1.2380337 , ...,  0.6036569 ,\n",
       "         -0.7110554 ,  0.5198825 ],\n",
       "        ...,\n",
       "        [ 0.02240079, -0.6947618 , -0.27486756, ...,  0.49084184,\n",
       "          0.8291764 ,  0.4097393 ],\n",
       "        [ 0.02478749, -0.64638567, -0.2850237 , ...,  0.47081646,\n",
       "          0.9248915 ,  0.43377233],\n",
       "        [ 0.01837321, -0.67970955, -0.30820352, ...,  0.5614533 ,\n",
       "          0.91682816,  0.46048862]]], dtype=float32)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
lucas_miranda's avatar
lucas_miranda committed
757
758
759
760
761
762
    }
   ],
   "source": [
    "gmvaep.predict(data_prep[:10])"
   ]
  },
763
764
  {
   "cell_type": "code",
lucas_miranda's avatar
lucas_miranda committed
765
   "execution_count": 15,
766
767
768
769
770
771
772
773
774
775
776
777
778
779
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Uncomment to see model summaries\n",
    "# encoder.summary()\n",
    "# decoder.summary()\n",
    "# grouper.summary()\n",
    "# gmvaep.summary()"
   ]
  },
  {
   "cell_type": "code",
lucas_miranda's avatar
lucas_miranda committed
780
   "execution_count": 16,
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
   "metadata": {},
   "outputs": [],
   "source": [
    "# Uncomment to plot model structure\n",
    "def plot_model(model, name):\n",
    "    tf.keras.utils.plot_model(\n",
    "        model,\n",
    "        to_file=os.path.join(\n",
    "            path,\n",
    "            \"deepof_{}_{}.png\".format(name, datetime.now().strftime(\"%Y%m%d-%H%M%S\")),\n",
    "        ),\n",
    "        show_shapes=True,\n",
    "        show_dtype=False,\n",
    "        show_layer_names=True,\n",
    "        rankdir=\"TB\",\n",
    "        expand_nested=True,\n",
    "        dpi=200,\n",
    "    )\n",
    "\n",
    "\n",
    "# plot_model(encoder, \"encoder\")\n",
    "# plot_model(decoder, \"decoder\")\n",
    "# plot_model(grouper, \"grouper\")\n",
    "# plot_model(gmvaep, \"gmvaep\")"
   ]
  },
lucas_miranda's avatar
lucas_miranda committed
807
808
809
810
811
812
813
814
815
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. Visualize priors"
   ]
  },
  {
   "cell_type": "code",
lucas_miranda's avatar
lucas_miranda committed
816
   "execution_count": 17,
lucas_miranda's avatar
lucas_miranda committed
817
818
819
820
821
822
823
824
825
826
827
828
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow_probability as tfp\n",
    "\n",
    "tfb = tfp.bijectors\n",
    "tfd = tfp.distributions\n",
    "tfpl = tfp.layers"
   ]
  },
  {
   "cell_type": "code",
lucas_miranda's avatar
lucas_miranda committed
829
   "execution_count": 18,
lucas_miranda's avatar
lucas_miranda committed
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_prior(number_of_components, encoding, init):\n",
    "\n",
    "    prior = tfd.MixtureSameFamily(\n",
    "        mixture_distribution=tfd.categorical.Categorical(\n",
    "            probs=tf.ones(number_of_components) / number_of_components\n",
    "        ),\n",
    "        components_distribution=tfd.MultivariateNormalDiag(\n",
    "            loc=tf.Variable(\n",
    "                init([number_of_components, encoding],),\n",
    "                name=\"prior_means\",\n",
    "            ),\n",
    "            scale_diag=tfp.util.TransformedVariable(\n",
    "                tf.ones([number_of_components, encoding]) / number_of_components,\n",
    "                tfb.Softplus(),\n",
    "                name=\"prior_scales\",\n",
    "            ),\n",
    "        ),\n",
    "    )\n",
    "    \n",
    "    return prior"
   ]
  },
  {
   "cell_type": "code",
lucas_miranda's avatar
lucas_miranda committed
857
   "execution_count": 19,
lucas_miranda's avatar
lucas_miranda committed
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
   "metadata": {},
   "outputs": [],
   "source": [
    "def sample_and_plot(prior, samples, ax, label):\n",
    "    \"\"\"Sample from the prior and plot with colours corresponding to different clusters\"\"\"\n",
    "    \n",
    "    samples = prior.sample(samples)\n",
    "    means = prior.components_distribution.mean()\n",
    "    samples = tf.concat([samples, means], axis=0)\n",
    "    pca = PCA(n_components=2)\n",
    "    prior = pca.fit_transform(samples)\n",
    "    \n",
    "    samples = prior[:-number_of_components, :]\n",
    "    means = prior[-number_of_components:, :]\n",
    "    \n",
    "    ax.scatter(prior[:,0], prior[:,1])\n",
    "    ax.scatter(means[:,0], means[:,1], label=label)\n",
    "    ax.set_xlabel(\"PC1\")\n",
    "    ax.set_ylabel(\"PC2\")\n",
    "    ax.legend()"
   ]
  },
lucas_miranda's avatar
lucas_miranda committed
880
881
  {
   "cell_type": "code",
882
   "execution_count": 29,
lucas_miranda's avatar
lucas_miranda committed
883
884
885
886
887
888
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
889
      "image/png": "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\n",
lucas_miranda's avatar
lucas_miranda committed
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
      "text/plain": [
       "<Figure size 720x1080 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "number_of_components = 15\n",
    "encoding = 6\n",
    "\n",
    "initializers = [\n",
    "    tf.keras.initializers.Constant(),\n",
    "    tf.keras.initializers.GlorotNormal(),\n",
    "    tf.keras.initializers.GlorotUniform(),\n",
    "    tf.keras.initializers.HeNormal(),\n",
    "    tf.keras.initializers.HeUniform(),\n",
    "    tf.keras.initializers.LecunNormal(),\n",
    "    tf.keras.initializers.LecunUniform(),\n",
    "    tf.keras.initializers.Orthogonal(),\n",
    "    tf.keras.initializers.RandomNormal(),\n",
    "    tf.keras.initializers.RandomUniform(),\n",
    "    tf.keras.initializers.TruncatedNormal(),\n",
    "    tf.keras.initializers.VarianceScaling(),\n",
    "]\n",
    "\n",
    "fig, ax = plt.subplots(4, 3, figsize=(10, 15), sharex=True, sharey=True)\n",
    "ax = [item for sublist in ax for item in sublist]\n",
    "\n",
    "for i, x in enumerate(ax):\n",
    "    prior = get_prior(number_of_components, encoding, initializers[i])\n",
    "    sample_and_plot(\n",
    "        prior,\n",
    "        1000,\n",
    "        x,\n",
    "        label=re.findall(\"initializers_v2.(.*?) \", str(initializers[i]))[0],\n",
    "    )\n",
    "\n",
    "fig.tight_layout(rect=[0.0, 0.0, 1.0, 0.97])\n",
    "plt.suptitle(\"effect of initialization on deepOF prior\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
939
   "execution_count": 31,
lucas_miranda's avatar
lucas_miranda committed
940
941
942
943
944
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
945
       "model_id": "0602dad9788142669a23ad9a2cd22d8c",
lucas_miranda's avatar
lucas_miranda committed
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(HTML(value=''), FloatProgress(value=0.0, max=12.0), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "from scipy.spatial.distance import pdist\n",
    "\n",
    "\n",
    "def quantify_separation(init, samples):\n",
    "\n",
    "    distances = []\n",
    "\n",
    "    for i in range(samples):\n",
    "        means = get_prior(\n",
    "            number_of_components, encoding, init\n",
    "        ).components_distribution.mean()\n",
    "        mean_dist = np.mean(pdist(means))\n",
    "        distances.append(mean_dist)\n",
    "\n",
    "    return (\n",
    "        np.mean(distances),\n",
    "        np.min(distances),\n",
    "        np.max(distances),\n",
    "        1.96 * np.std(distances),\n",
    "    )\n",
    "\n",
    "\n",
    "prior_init_eval_dict = {}\n",
    "for init in tqdm.tqdm(initializers):\n",
    "    prior_init_eval_dict[\n",
    "        re.findall(\"initializers_v2.(.*?) \", str(init))[0]\n",
991
    "    ] = quantify_separation(init, 100)"
lucas_miranda's avatar
lucas_miranda committed
992
993
994
995
   ]
  },
  {
   "cell_type": "code",
996
   "execution_count": 32,
lucas_miranda's avatar
lucas_miranda committed
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>min</th>\n",
       "      <th>max</th>\n",
       "      <th>CI95</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>HeUniform</th>\n",
1029
1030
1031
1032
       "      <td>1.239522</td>\n",
       "      <td>1.094515</td>\n",
       "      <td>1.382325</td>\n",
       "      <td>0.118654</td>\n",
lucas_miranda's avatar
lucas_miranda committed
1033
1034
1035
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HeNormal</th>\n",
1036
1037
1038
1039
       "      <td>1.227263</td>\n",
       "      <td>1.005923</td>\n",
       "      <td>1.430978</td>\n",
       "      <td>0.147760</td>\n",
lucas_miranda's avatar
lucas_miranda committed
1040
1041
1042
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GlorotUniform</th>\n",
1043
1044
1045
1046
       "      <td>1.033926</td>\n",
       "      <td>0.879322</td>\n",
       "      <td>1.165203</td>\n",
       "      <td>0.118878</td>\n",
lucas_miranda's avatar
lucas_miranda committed
1047
1048
1049
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GlorotNormal</th>\n",
1050
1051
1052
1053
       "      <td>1.033110</td>\n",
       "      <td>0.811878</td>\n",
       "      <td>1.211693</td>\n",
       "      <td>0.148522</td>\n",
lucas_miranda's avatar
lucas_miranda committed
1054
1055
       "    </tr>\n",
       "    <tr>\n",
1056
1057
1058
1059
1060
       "      <th>LecunUniform</th>\n",
       "      <td>0.874425</td>\n",
       "      <td>0.727023</td>\n",
       "      <td>0.982678</td>\n",
       "      <td>0.087684</td>\n",
lucas_miranda's avatar
lucas_miranda committed
1061
1062
       "    </tr>\n",
       "    <tr>\n",
1063
1064
1065
1066
1067
       "      <th>Orthogonal</th>\n",
       "      <td>0.873795</td>\n",
       "      <td>0.840862</td>\n",
       "      <td>0.906167</td>\n",
       "      <td>0.025545</td>\n",
lucas_miranda's avatar
lucas_miranda committed
1068
1069
1070
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VarianceScaling</th>\n",
1071
1072
1073
1074
       "      <td>0.865526</td>\n",
       "      <td>0.710835</td>\n",
       "      <td>0.988492</td>\n",
       "      <td>0.106752</td>\n",
lucas_miranda's avatar
lucas_miranda committed
1075
1076
1077
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LecunNormal</th>\n",
1078
1079
1080
1081
       "      <td>0.864939</td>\n",
       "      <td>0.726335</td>\n",
       "      <td>1.020095</td>\n",
       "      <td>0.109755</td>\n",
lucas_miranda's avatar
lucas_miranda committed
1082
1083
1084
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RandomNormal</th>\n",
1085
1086
1087
1088
       "      <td>0.164709</td>\n",
       "      <td>0.139836</td>\n",
       "      <td>0.199212</td>\n",
       "      <td>0.023759</td>\n",
lucas_miranda's avatar
lucas_miranda committed
1089
1090
1091
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TruncatedNormal</th>\n",
1092
1093
1094
1095
       "      <td>0.145291</td>\n",
       "      <td>0.113060</td>\n",
       "      <td>0.166927</td>\n",
       "      <td>0.019680</td>\n",
lucas_miranda's avatar
lucas_miranda committed
1096
1097
1098
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RandomUniform</th>\n",
1099
1100
1101
1102
       "      <td>0.097162</td>\n",
       "      <td>0.083930</td>\n",
       "      <td>0.111233</td>\n",
       "      <td>0.010793</td>\n",
lucas_miranda's avatar
lucas_miranda committed
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Constant</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     mean       min       max      CI95\n",
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
       "HeUniform        1.239522  1.094515  1.382325  0.118654\n",
       "HeNormal         1.227263  1.005923  1.430978  0.147760\n",
       "GlorotUniform    1.033926  0.879322  1.165203  0.118878\n",
       "GlorotNormal     1.033110  0.811878  1.211693  0.148522\n",
       "LecunUniform     0.874425  0.727023  0.982678  0.087684\n",
       "Orthogonal       0.873795  0.840862  0.906167  0.025545\n",
       "VarianceScaling  0.865526  0.710835  0.988492  0.106752\n",
       "LecunNormal      0.864939  0.726335  1.020095  0.109755\n",
       "RandomNormal     0.164709  0.139836  0.199212  0.023759\n",
       "TruncatedNormal  0.145291  0.113060  0.166927  0.019680\n",
       "RandomUniform    0.097162  0.083930  0.111233  0.010793\n",
lucas_miranda's avatar
lucas_miranda committed
1128
1129
1130
       "Constant         0.000000  0.000000  0.000000  0.000000"
      ]
     },
1131
     "execution_count": 32,
lucas_miranda's avatar
lucas_miranda committed
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prior_init_eval = pd.DataFrame(prior_init_eval_dict).T\n",
    "prior_init_eval.rename(columns={0:\"mean\", 1:\"min\", 2:\"max\", 3:\"CI95\"}, inplace=True)\n",
    "prior_init_eval.sort_values(\"mean\", ascending=False)"
   ]
  },
1142
1143
1144
1145
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
1146
    "### 4. Evaluate reconstruction (to be incorporated into deepof.evaluate)"
1147
1148
1149
1150
   ]
  },
  {
   "cell_type": "code",
1151
   "execution_count": 191,
1152
1153
1154
   "metadata": {},
   "outputs": [],
   "source": [
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
    "# Auxiliary animation functions\n",
    "\n",
    "\n",
    "def plot_mouse_graph(instant_x, instant_y, instant_rec_x, instant_rec_y, ax, edges):\n",
    "    \"\"\"Generates a graph plot of the mouse\"\"\"\n",
    "    plots = []\n",
    "    rec_plots = []\n",
    "    for edge in edges:\n",
    "        (temp_plot,) = ax.plot(\n",
    "            [float(instant_x[edge[0]]), float(instant_x[edge[1]])],\n",
    "            [float(instant_y[edge[0]]), float(instant_y[edge[1]])],\n",
    "            color=\"#006699\",\n",
1167
    "            linewidth=2.0,\n",
1168
1169
1170
1171
    "        )\n",
    "        (temp_rec_plot,) = ax.plot(\n",
    "            [float(instant_rec_x[edge[0]]), float(instant_rec_x[edge[1]])],\n",
    "            [float(instant_rec_y[edge[0]]), float(instant_rec_y[edge[1]])],\n",
1172
1173
    "            color=\"red\",\n",
    "            linewidth=2.0,\n",
1174
1175
1176
    "        )\n",
    "        plots.append(temp_plot)\n",
    "        rec_plots.append(temp_rec_plot)\n",
1177
    "    return plots, rec_plots\n",
1178
1179
    "\n",
    "\n",
1180
    "def update_mouse_graph(x, y, rec_x, rec_y, plots, rec_plots, edges):\n",
1181
1182
1183
1184
1185
1186
    "    \"\"\"Updates the graph plot to enable animation\"\"\"\n",
    "\n",
    "    for plot, edge in zip(plots, edges):\n",
    "        plot.set_data(\n",
    "            [float(x[edge[0]]), float(x[edge[1]])],\n",
    "            [float(y[edge[0]]), float(y[edge[1]])],\n",
1187
1188
1189
1190
1191
    "        )\n",
    "    for plot, edge in zip(rec_plots, edges):\n",
    "        plot.set_data(\n",
    "            [float(rec_x[edge[0]]), float(rec_x[edge[1]])],\n",
    "            [float(rec_y[edge[0]]), float(rec_y[edge[1]])],\n",
1192
1193
1194
1195
1196
    "        )"
   ]
  },
  {
   "cell_type": "code",
1197
   "execution_count": 192,
1198
1199
1200
   "metadata": {
    "scrolled": false
   },
1201
1202
1203
1204
1205
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
1206
      "Test 11_s11\n"
1207
1208
1209
1210
     ]
    },
    {
     "data": {
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717