Mercurial > repos > goeckslab > ludwig_render_config

comparison test-data/temp_model01/model_hyperparameters.json @ 0:ed8a9ea5bc73 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/goeckslab/Galaxy-Ludwig.git commit bdea9430787658783a51cc6c2ae951a01e455bb4
author goeckslab
date Tue, 07 Jan 2025 22:45:58 +0000
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:ed8a9ea5bc73
1 {
2 "backend": null,
3 "combiner": {
4 "activation": "relu",
5 "bias_initializer": "zeros",
6 "dropout": 0.0,
7 "fc_layers": null,
8 "flatten_inputs": false,
9 "norm": null,
10 "norm_params": null,
11 "num_fc_layers": 0,
12 "output_size": 256,
13 "residual": false,
14 "type": "concat",
15 "use_bias": true,
16 "weights_initializer": "xavier_uniform"
17 },
18 "defaults": {
19 "audio": {
20 "encoder": {
21 "activation": "relu",
22 "bias_initializer": "zeros",
23 "conv_layers": null,
24 "dropout": 0.0,
25 "embedding_size": 256,
26 "embeddings_on_cpu": false,
27 "embeddings_trainable": true,
28 "fc_layers": null,
29 "filter_size": 3,
30 "max_sequence_length": null,
31 "norm": null,
32 "norm_params": null,
33 "num_conv_layers": null,
34 "num_fc_layers": null,
35 "num_filters": 256,
36 "output_size": 256,
37 "pool_function": "max",
38 "pool_size": null,
39 "pretrained_embeddings": null,
40 "reduce_output": "sum",
41 "representation": "dense",
42 "should_embed": true,
43 "skip": false,
44 "type": "parallel_cnn",
45 "use_bias": true,
46 "vocab": null,
47 "weights_initializer": "xavier_uniform"
48 },
49 "preprocessing": {
50 "audio_file_length_limit_in_s": 7.5,
51 "computed_fill_value": null,
52 "fill_value": null,
53 "in_memory": true,
54 "missing_value_strategy": "bfill",
55 "norm": null,
56 "num_fft_points": null,
57 "num_filter_bands": 80,
58 "padding_value": 0.0,
59 "type": "fbank",
60 "window_length_in_s": 0.04,
61 "window_shift_in_s": 0.02,
62 "window_type": "hamming"
63 }
64 },
65 "bag": {
66 "encoder": {
67 "activation": "relu",
68 "bias_initializer": "zeros",
69 "dropout": 0.0,
70 "embedding_size": 50,
71 "embeddings_on_cpu": false,
72 "embeddings_trainable": true,
73 "fc_layers": null,
74 "force_embedding_size": false,
75 "norm": null,
76 "norm_params": null,
77 "num_fc_layers": 0,
78 "output_size": 10,
79 "pretrained_embeddings": null,
80 "representation": "dense",
81 "skip": false,
82 "type": "embed",
83 "use_bias": true,
84 "vocab": null,
85 "weights_initializer": "xavier_uniform"
86 },
87 "preprocessing": {
88 "computed_fill_value": "<UNK>",
89 "fill_value": "<UNK>",
90 "lowercase": false,
91 "missing_value_strategy": "fill_with_const",
92 "most_common": 10000,
93 "tokenizer": "space"
94 }
95 },
96 "binary": {
97 "decoder": {
98 "bias_initializer": "zeros",
99 "fc_activation": "relu",
100 "fc_bias_initializer": "zeros",
101 "fc_dropout": 0.0,
102 "fc_layers": null,
103 "fc_norm": null,
104 "fc_norm_params": null,
105 "fc_output_size": 256,
106 "fc_use_bias": true,
107 "fc_weights_initializer": "xavier_uniform",
108 "input_size": null,
109 "num_fc_layers": 0,
110 "type": "regressor",
111 "use_bias": true,
112 "weights_initializer": "xavier_uniform"
113 },
114 "encoder": {
115 "skip": false,
116 "type": "passthrough"
117 },
118 "loss": {
119 "confidence_penalty": 0,
120 "positive_class_weight": null,
121 "robust_lambda": 0,
122 "type": "binary_weighted_cross_entropy",
123 "weight": 1.0
124 },
125 "preprocessing": {
126 "computed_fill_value": null,
127 "fallback_true_label": null,
128 "fill_value": null,
129 "missing_value_strategy": "fill_with_false"
130 }
131 },
132 "category": {
133 "decoder": {
134 "bias_initializer": "zeros",
135 "fc_activation": "relu",
136 "fc_bias_initializer": "zeros",
137 "fc_dropout": 0.0,
138 "fc_layers": null,
139 "fc_norm": null,
140 "fc_norm_params": null,
141 "fc_output_size": 256,
142 "fc_use_bias": true,
143 "fc_weights_initializer": "xavier_uniform",
144 "input_size": null,
145 "num_classes": null,
146 "num_fc_layers": 0,
147 "type": "classifier",
148 "use_bias": true,
149 "weights_initializer": "xavier_uniform"
150 },
151 "encoder": {
152 "dropout": 0.0,
153 "embedding_initializer": null,
154 "embedding_size": 50,
155 "embeddings_on_cpu": false,
156 "embeddings_trainable": true,
157 "pretrained_embeddings": null,
158 "skip": false,
159 "type": "dense",
160 "vocab": null
161 },
162 "loss": {
163 "class_similarities": null,
164 "class_similarities_temperature": 0,
165 "class_weights": null,
166 "confidence_penalty": 0,
167 "robust_lambda": 0,
168 "type": "softmax_cross_entropy",
169 "weight": 1.0
170 },
171 "preprocessing": {
172 "cache_encoder_embeddings": false,
173 "computed_fill_value": "<UNK>",
174 "fill_value": "<UNK>",
175 "lowercase": false,
176 "missing_value_strategy": "fill_with_const",
177 "most_common": 10000
178 }
179 },
180 "date": {
181 "encoder": {
182 "activation": "relu",
183 "bias_initializer": "zeros",
184 "dropout": 0.0,
185 "embedding_size": 10,
186 "embeddings_on_cpu": false,
187 "fc_layers": null,
188 "norm": null,
189 "norm_params": null,
190 "num_fc_layers": 0,
191 "output_size": 10,
192 "skip": false,
193 "type": "embed",
194 "use_bias": true,
195 "weights_initializer": "xavier_uniform"
196 },
197 "preprocessing": {
198 "computed_fill_value": "",
199 "datetime_format": null,
200 "fill_value": "",
201 "missing_value_strategy": "fill_with_const"
202 }
203 },
204 "h3": {
205 "encoder": {
206 "activation": "relu",
207 "bias_initializer": "zeros",
208 "dropout": 0.0,
209 "embedding_size": 10,
210 "embeddings_on_cpu": false,
211 "fc_layers": null,
212 "norm": null,
213 "norm_params": null,
214 "num_fc_layers": 0,
215 "output_size": 10,
216 "reduce_output": "sum",
217 "skip": false,
218 "type": "embed",
219 "use_bias": true,
220 "weights_initializer": "xavier_uniform"
221 },
222 "preprocessing": {
223 "computed_fill_value": 576495936675512319,
224 "fill_value": 576495936675512319,
225 "missing_value_strategy": "fill_with_const"
226 }
227 },
228 "image": {
229 "augmentation": [],
230 "decoder": {
231 "conv_norm": "batch",
232 "fc_activation": "relu",
233 "fc_bias_initializer": "zeros",
234 "fc_dropout": 0.0,
235 "fc_layers": null,
236 "fc_norm": null,
237 "fc_norm_params": null,
238 "fc_output_size": 256,
239 "fc_use_bias": true,
240 "fc_weights_initializer": "xavier_uniform",
241 "height": null,
242 "input_size": 1024,
243 "num_channels": null,
244 "num_classes": null,
245 "num_fc_layers": 0,
246 "type": "unet",
247 "width": null
248 },
249 "encoder": {
250 "conv_activation": "relu",
251 "conv_dropout": 0.0,
252 "conv_layers": null,
253 "conv_norm": null,
254 "conv_norm_params": null,
255 "conv_use_bias": true,
256 "dilation": 1,
257 "fc_activation": "relu",
258 "fc_bias_initializer": "zeros",
259 "fc_dropout": 0.0,
260 "fc_layers": null,
261 "fc_norm": null,
262 "fc_norm_params": null,
263 "fc_use_bias": true,
264 "fc_weights_initializer": "xavier_uniform",
265 "groups": 1,
266 "height": null,
267 "kernel_size": 3,
268 "num_channels": null,
269 "num_conv_layers": null,
270 "num_fc_layers": 1,
271 "out_channels": 32,
272 "output_size": 128,
273 "padding": "valid",
274 "padding_mode": "zeros",
275 "pool_dilation": 1,
276 "pool_function": "max",
277 "pool_kernel_size": 2,
278 "pool_padding": 0,
279 "pool_stride": null,
280 "skip": false,
281 "stride": 1,
282 "type": "stacked_cnn",
283 "width": null
284 },
285 "loss": {
286 "class_similarities": null,
287 "class_similarities_temperature": 0,
288 "class_weights": null,
289 "confidence_penalty": 0,
290 "robust_lambda": 0,
291 "type": "softmax_cross_entropy",
292 "weight": 1.0
293 },
294 "preprocessing": {
295 "computed_fill_value": null,
296 "fill_value": null,
297 "height": null,
298 "in_memory": true,
299 "infer_image_dimensions": true,
300 "infer_image_max_height": 256,
301 "infer_image_max_width": 256,
302 "infer_image_num_channels": true,
303 "infer_image_num_classes": false,
304 "infer_image_sample_size": 100,
305 "missing_value_strategy": "bfill",
306 "num_channels": null,
307 "num_classes": null,
308 "num_processes": 1,
309 "requires_equal_dimensions": false,
310 "resize_method": "interpolate",
311 "standardize_image": null,
312 "width": null
313 }
314 },
315 "number": {
316 "decoder": {
317 "bias_initializer": "zeros",
318 "fc_activation": "relu",
319 "fc_bias_initializer": "zeros",
320 "fc_dropout": 0.0,
321 "fc_layers": null,
322 "fc_norm": null,
323 "fc_norm_params": null,
324 "fc_output_size": 256,
325 "fc_use_bias": true,
326 "fc_weights_initializer": "xavier_uniform",
327 "input_size": null,
328 "num_fc_layers": 0,
329 "type": "regressor",
330 "use_bias": true,
331 "weights_initializer": "xavier_uniform"
332 },
333 "encoder": {
334 "skip": false,
335 "type": "passthrough"
336 },
337 "loss": {
338 "type": "mean_squared_error",
339 "weight": 1.0
340 },
341 "preprocessing": {
342 "computed_fill_value": 0.0,
343 "computed_outlier_fill_value": 0.0,
344 "fill_value": 0.0,
345 "missing_value_strategy": "fill_with_const",
346 "normalization": "zscore",
347 "outlier_strategy": null,
348 "outlier_threshold": 3.0
349 }
350 },
351 "sequence": {
352 "decoder": {
353 "cell_type": "gru",
354 "fc_activation": "relu",
355 "fc_bias_initializer": "zeros",
356 "fc_dropout": 0.0,
357 "fc_layers": null,
358 "fc_norm": null,
359 "fc_norm_params": null,
360 "fc_output_size": 256,
361 "fc_use_bias": true,
362 "fc_weights_initializer": "xavier_uniform",
363 "input_size": 256,
364 "max_sequence_length": null,
365 "num_fc_layers": 0,
366 "num_layers": 1,
367 "reduce_input": "sum",
368 "type": "generator",
369 "vocab_size": null
370 },
371 "encoder": {
372 "dropout": 0.0,
373 "embedding_size": 256,
374 "embeddings_on_cpu": false,
375 "embeddings_trainable": true,
376 "max_sequence_length": null,
377 "pretrained_embeddings": null,
378 "reduce_output": "sum",
379 "representation": "dense",
380 "skip": false,
381 "type": "embed",
382 "vocab": null,
383 "weights_initializer": "uniform"
384 },
385 "loss": {
386 "class_similarities": null,
387 "class_similarities_temperature": 0,
388 "class_weights": null,
389 "confidence_penalty": 0,
390 "robust_lambda": 0,
391 "type": "sequence_softmax_cross_entropy",
392 "unique": false,
393 "weight": 1.0
394 },
395 "preprocessing": {
396 "cache_encoder_embeddings": false,
397 "computed_fill_value": "<UNK>",
398 "fill_value": "<UNK>",
399 "lowercase": false,
400 "max_sequence_length": 256,
401 "missing_value_strategy": "fill_with_const",
402 "most_common": 20000,
403 "ngram_size": 2,
404 "padding": "right",
405 "padding_symbol": "<PAD>",
406 "sequence_length": null,
407 "tokenizer": "space",
408 "unknown_symbol": "<UNK>",
409 "vocab_file": null
410 }
411 },
412 "set": {
413 "decoder": {
414 "bias_initializer": "zeros",
415 "fc_activation": "relu",
416 "fc_bias_initializer": "zeros",
417 "fc_dropout": 0.0,
418 "fc_layers": null,
419 "fc_norm": null,
420 "fc_norm_params": null,
421 "fc_output_size": 256,
422 "fc_use_bias": true,
423 "fc_weights_initializer": "xavier_uniform",
424 "input_size": null,
425 "num_classes": null,
426 "num_fc_layers": 0,
427 "type": "classifier",
428 "use_bias": true,
429 "weights_initializer": "xavier_uniform"
430 },
431 "encoder": {
432 "activation": "relu",
433 "bias_initializer": "zeros",
434 "dropout": 0.0,
435 "embedding_size": 50,
436 "embeddings_on_cpu": false,
437 "embeddings_trainable": true,
438 "fc_layers": null,
439 "norm": null,
440 "norm_params": null,
441 "num_fc_layers": 0,
442 "output_size": 10,
443 "pretrained_embeddings": null,
444 "representation": "dense",
445 "skip": false,
446 "type": "embed",
447 "use_bias": true,
448 "vocab": null,
449 "weights_initializer": "xavier_uniform"
450 },
451 "loss": {
452 "class_weights": null,
453 "type": "sigmoid_cross_entropy",
454 "weight": 1.0
455 },
456 "preprocessing": {
457 "computed_fill_value": "<UNK>",
458 "fill_value": "<UNK>",
459 "lowercase": false,
460 "missing_value_strategy": "fill_with_const",
461 "most_common": 10000,
462 "tokenizer": "space"
463 }
464 },
465 "text": {
466 "decoder": {
467 "cell_type": "gru",
468 "fc_activation": "relu",
469 "fc_bias_initializer": "zeros",
470 "fc_dropout": 0.0,
471 "fc_layers": null,
472 "fc_norm": null,
473 "fc_norm_params": null,
474 "fc_output_size": 256,
475 "fc_use_bias": true,
476 "fc_weights_initializer": "xavier_uniform",
477 "input_size": 256,
478 "max_sequence_length": null,
479 "num_fc_layers": 0,
480 "num_layers": 1,
481 "reduce_input": "sum",
482 "type": "generator",
483 "vocab_size": null
484 },
485 "encoder": {
486 "activation": "relu",
487 "bias_initializer": "zeros",
488 "conv_layers": null,
489 "dropout": 0.0,
490 "embedding_size": 256,
491 "embeddings_on_cpu": false,
492 "embeddings_trainable": true,
493 "fc_layers": null,
494 "filter_size": 3,
495 "max_sequence_length": null,
496 "norm": null,
497 "norm_params": null,
498 "num_conv_layers": null,
499 "num_fc_layers": null,
500 "num_filters": 256,
501 "output_size": 256,
502 "pool_function": "max",
503 "pool_size": null,
504 "pretrained_embeddings": null,
505 "reduce_output": "sum",
506 "representation": "dense",
507 "should_embed": true,
508 "skip": false,
509 "type": "parallel_cnn",
510 "use_bias": true,
511 "vocab": null,
512 "weights_initializer": "xavier_uniform"
513 },
514 "loss": {
515 "class_similarities": null,
516 "class_similarities_temperature": 0,
517 "class_weights": null,
518 "confidence_penalty": 0,
519 "robust_lambda": 0,
520 "type": "sequence_softmax_cross_entropy",
521 "unique": false,
522 "weight": 1.0
523 },
524 "preprocessing": {
525 "cache_encoder_embeddings": false,
526 "compute_idf": false,
527 "computed_fill_value": "<UNK>",
528 "fill_value": "<UNK>",
529 "lowercase": false,
530 "max_sequence_length": 256,
531 "missing_value_strategy": "fill_with_const",
532 "most_common": 20000,
533 "ngram_size": 2,
534 "padding": "right",
535 "padding_symbol": "<PAD>",
536 "pretrained_model_name_or_path": null,
537 "prompt": {
538 "retrieval": {
539 "index_name": null,
540 "k": 0,
541 "model_name": null,
542 "type": null
543 },
544 "task": null,
545 "template": null
546 },
547 "sequence_length": null,
548 "tokenizer": "space_punct",
549 "unknown_symbol": "<UNK>",
550 "vocab_file": null
551 }
552 },
553 "timeseries": {
554 "decoder": {
555 "activation": null,
556 "bias_initializer": "zeros",
557 "clip": null,
558 "fc_activation": "relu",
559 "fc_bias_initializer": "zeros",
560 "fc_dropout": 0.0,
561 "fc_layers": null,
562 "fc_norm": null,
563 "fc_norm_params": null,
564 "fc_output_size": 256,
565 "fc_use_bias": true,
566 "fc_weights_initializer": "xavier_uniform",
567 "input_size": null,
568 "multiplier": 1.0,
569 "num_fc_layers": 0,
570 "output_size": null,
571 "type": "projector",
572 "use_bias": true,
573 "weights_initializer": "xavier_uniform"
574 },
575 "encoder": {
576 "activation": "relu",
577 "bias_initializer": "zeros",
578 "conv_layers": null,
579 "dropout": 0.0,
580 "embedding_size": 256,
581 "embeddings_on_cpu": false,
582 "embeddings_trainable": true,
583 "fc_layers": null,
584 "filter_size": 3,
585 "max_sequence_length": null,
586 "norm": null,
587 "norm_params": null,
588 "num_conv_layers": null,
589 "num_fc_layers": null,
590 "num_filters": 256,
591 "output_size": 256,
592 "pool_function": "max",
593 "pool_size": null,
594 "pretrained_embeddings": null,
595 "reduce_output": "sum",
596 "representation": "dense",
597 "should_embed": true,
598 "skip": false,
599 "type": "parallel_cnn",
600 "use_bias": true,
601 "vocab": null,
602 "weights_initializer": "xavier_uniform"
603 },
604 "loss": {
605 "delta": 1.0,
606 "type": "huber",
607 "weight": 1.0
608 },
609 "preprocessing": {
610 "computed_fill_value": "",
611 "fill_value": "",
612 "missing_value_strategy": "fill_with_const",
613 "padding": "right",
614 "padding_value": 0.0,
615 "timeseries_length_limit": 256,
616 "tokenizer": "space",
617 "window_size": 0
618 }
619 },
620 "vector": {
621 "decoder": {
622 "activation": null,
623 "bias_initializer": "zeros",
624 "clip": null,
625 "fc_activation": "relu",
626 "fc_bias_initializer": "zeros",
627 "fc_dropout": 0.0,
628 "fc_layers": null,
629 "fc_norm": null,
630 "fc_norm_params": null,
631 "fc_output_size": 256,
632 "fc_use_bias": true,
633 "fc_weights_initializer": "xavier_uniform",
634 "input_size": null,
635 "multiplier": 1.0,
636 "num_fc_layers": 0,
637 "output_size": null,
638 "type": "projector",
639 "use_bias": true,
640 "weights_initializer": "xavier_uniform"
641 },
642 "encoder": {
643 "activation": "relu",
644 "bias_initializer": "zeros",
645 "dropout": 0.0,
646 "fc_layers": null,
647 "input_size": null,
648 "norm": null,
649 "norm_params": null,
650 "num_layers": 1,
651 "output_size": 256,
652 "skip": false,
653 "type": "dense",
654 "use_bias": true,
655 "weights_initializer": "xavier_uniform"
656 },
657 "loss": {
658 "type": "mean_squared_error",
659 "weight": 1.0
660 },
661 "preprocessing": {
662 "computed_fill_value": "",
663 "fill_value": "",
664 "missing_value_strategy": "fill_with_const",
665 "vector_size": null
666 }
667 }
668 },
669 "hyperopt": null,
670 "input_features": [
671 {
672 "active": true,
673 "column": "temperature_feature",
674 "encoder": {
675 "activation": "tanh",
676 "bias_initializer": "zeros",
677 "bidirectional": false,
678 "cell_type": "rnn",
679 "dropout": 0.0,
680 "embedding_size": 1,
681 "embeddings_on_cpu": false,
682 "embeddings_trainable": true,
683 "fc_activation": "relu",
684 "fc_dropout": 0.0,
685 "fc_layers": null,
686 "input_size": 20,
687 "max_sequence_length": 20,
688 "norm": null,
689 "norm_params": null,
690 "num_fc_layers": 0,
691 "num_layers": 1,
692 "output_size": 256,
693 "pretrained_embeddings": null,
694 "recurrent_activation": "sigmoid",
695 "recurrent_dropout": 0.0,
696 "recurrent_initializer": "orthogonal",
697 "reduce_output": "last",
698 "representation": "dense",
699 "saved_weights_in_checkpoint": true,
700 "should_embed": false,
701 "skip": false,
702 "state_size": 32,
703 "type": "rnn",
704 "unit_forget_bias": true,
705 "use_bias": true,
706 "vocab": null,
707 "weights_initializer": "xavier_uniform"
708 },
709 "name": "temperature_feature",
710 "preprocessing": {
711 "cache_encoder_embeddings": false,
712 "computed_fill_value": "",
713 "fill_value": "",
714 "missing_value_strategy": "fill_with_const",
715 "padding": "right",
716 "padding_value": 0.0,
717 "timeseries_length_limit": 256,
718 "tokenizer": "space",
719 "window_size": 0
720 },
721 "proc_column": "temperature_feature_XJMv7q",
722 "tied": null,
723 "type": "timeseries"
724 }
725 ],
726 "ludwig_version": "0.10.1",
727 "model_type": "ecd",
728 "output_features": [
729 {
730 "active": true,
731 "clip": null,
732 "column": "temperature",
733 "decoder": {
734 "bias_initializer": "zeros",
735 "fc_activation": "relu",
736 "fc_bias_initializer": "zeros",
737 "fc_dropout": 0.0,
738 "fc_layers": null,
739 "fc_norm": null,
740 "fc_norm_params": null,
741 "fc_output_size": 256,
742 "fc_use_bias": true,
743 "fc_weights_initializer": "xavier_uniform",
744 "input_size": 32,
745 "num_fc_layers": 0,
746 "type": "regressor",
747 "use_bias": true,
748 "weights_initializer": "xavier_uniform"
749 },
750 "default_validation_metric": "mean_squared_error",
751 "dependencies": [],
752 "input_size": 32,
753 "loss": {
754 "type": "mean_squared_error",
755 "weight": 1.0
756 },
757 "name": "temperature",
758 "num_classes": null,
759 "preprocessing": {
760 "computed_fill_value": 0.0,
761 "computed_outlier_fill_value": 0.0,
762 "fill_value": 0.0,
763 "missing_value_strategy": "drop_row",
764 "normalization": null,
765 "outlier_strategy": null,
766 "outlier_threshold": 3.0
767 },
768 "proc_column": "temperature_VO1XAD",
769 "reduce_dependencies": "sum",
770 "reduce_input": "sum",
771 "type": "number"
772 }
773 ],
774 "preprocessing": {
775 "global_max_sequence_length": null,
776 "oversample_minority": null,
777 "sample_ratio": 1.0,
778 "sample_size": null,
779 "split": {
780 "probabilities": [
781 0.7,
782 0.1,
783 0.2
784 ],
785 "type": "random"
786 },
787 "undersample_majority": null
788 },
789 "trainer": {
790 "batch_size": 8,
791 "bucketing_field": null,
792 "checkpoints_per_epoch": 0,
793 "compile": false,
794 "early_stop": 5,
795 "effective_batch_size": "auto",
796 "enable_gradient_checkpointing": false,
797 "enable_profiling": false,
798 "epochs": 2,
799 "eval_batch_size": 128,
800 "eval_steps": null,
801 "evaluate_training_set": false,
802 "gradient_accumulation_steps": 1,
803 "gradient_clipping": {
804 "clipglobalnorm": 0.5,
805 "clipnorm": null,
806 "clipvalue": null
807 },
808 "increase_batch_size_eval_metric": "loss",
809 "increase_batch_size_eval_split": "training",
810 "increase_batch_size_on_plateau": 0,
811 "increase_batch_size_on_plateau_patience": 5,
812 "increase_batch_size_on_plateau_rate": 2.0,
813 "learning_rate": 0.001,
814 "learning_rate_scaling": "linear",
815 "learning_rate_scheduler": {
816 "decay": null,
817 "decay_rate": 0.96,
818 "decay_steps": 10000,
819 "eta_min": 0,
820 "reduce_eval_metric": "loss",
821 "reduce_eval_split": "training",
822 "reduce_on_plateau": 0,
823 "reduce_on_plateau_patience": 10,
824 "reduce_on_plateau_rate": 0.1,
825 "staircase": false,
826 "t_0": null,
827 "t_mult": 1,
828 "warmup_evaluations": 0,
829 "warmup_fraction": 0.0
830 },
831 "max_batch_size": 1099511627776,
832 "optimizer": {
833 "amsgrad": false,
834 "betas": [
835 0.9,
836 0.999
837 ],
838 "eps": 1e-08,
839 "type": "adam",
840 "weight_decay": 0.0
841 },
842 "profiler": {
843 "active": 3,
844 "repeat": 5,
845 "skip_first": 0,
846 "wait": 1,
847 "warmup": 1
848 },
849 "regularization_lambda": 0.0,
850 "regularization_type": "l2",
851 "should_shuffle": true,
852 "skip_all_evaluation": false,
853 "steps_per_checkpoint": 0,
854 "train_steps": null,
855 "use_mixed_precision": false,
856 "validation_field": "temperature",
857 "validation_metric": "mean_squared_error"
858 }
859 }