train_utils.py 17.4 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
# @author lucasmiranda42
# encoding: utf-8
# module deepof

"""

Simple utility functions used in deepof example scripts. These are not part of the main package

"""

11
from datetime import date, datetime
12
from kerastuner import BayesianOptimization, Hyperband
13
from kerastuner import HyperParameters
14
from kerastuner_tensorboard_logger import TensorBoardLogger
15
from tensorboard.plugins.hparams import api as hp
16
from typing import Tuple, Union, Any, List
17
18
19
20
21
22
23
import deepof.hypermodels
import deepof.model_utils
import numpy as np
import os
import pickle
import tensorflow as tf

24
25
26
27
28
# Ignore warning with no downstream effect
tf.get_logger().setLevel("ERROR")
tf.autograph.set_verbosity(0)


29
class CustomStopper(tf.keras.callbacks.EarlyStopping):
30
    """ Custom early stopping callback. Prevents the model from stopping before warmup is over """
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47

    def __init__(self, start_epoch, *args, **kwargs):
        super(CustomStopper, self).__init__(*args, **kwargs)
        self.start_epoch = start_epoch

    def get_config(self):  # pragma: no cover
        """Updates callback metadata"""

        config = super().get_config().copy()
        config.update({"start_epoch": self.start_epoch})
        return config

    def on_epoch_end(self, epoch, logs=None):
        if epoch > self.start_epoch:
            super().on_epoch_end(epoch, logs)


48
def load_hparams(hparams):
49
50
51
52
53
54
55
56
    """Loads hyperparameters from a custom dictionary pickled on disc.
    Thought to be used with the output of hyperparameter_tuning.py"""

    if hparams is not None:
        with open(hparams, "rb") as handle:
            hparams = pickle.load(handle)
    else:
        hparams = {
57
58
59
60
61
            "bidirectional_merge": "ave",
            "clipvalue": 1.0,
            "dense_activation": "relu",
            "dense_layers_per_branch": 1,
            "dropout_rate": 1e-3,
62
            "learning_rate": 1e-3,
63
64
65
            "units_conv": 160,
            "units_dense2": 120,
            "units_lstm": 300,
66
67
68
69
70
71
72
73
74
75
76
        }
    return hparams


def load_treatments(train_path):
    """Loads a dictionary containing the treatments per individual,
    to be loaded as metadata in the coordinates class"""
    try:
        with open(
            os.path.join(
                train_path,
77
                [i for i in os.listdir(train_path) if i.endswith(".pkl")][0],
78
79
80
81
82
83
84
85
86
87
88
            ),
            "rb",
        ) as handle:
            treatment_dict = pickle.load(handle)
    except IndexError:
        treatment_dict = None

    return treatment_dict


def get_callbacks(
lucas_miranda's avatar
lucas_miranda committed
89
90
91
92
    X_train: np.array,
    batch_size: int,
    cp: bool,
    variational: bool,
93
    phenotype_class: float,
lucas_miranda's avatar
lucas_miranda committed
94
95
    predictor: float,
    loss: str,
96
    logparam: dict = None,
97
    outpath: str = ".",
98
) -> List[Union[Any]]:
99
    """Generates callbacks for model training, including:
100
101
102
103
    - run_ID: run name, with coarse parameter details;
    - tensorboard_callback: for real-time visualization;
    - cp_callback: for checkpoint saving,
    - onecycle: for learning rate scheduling"""
104

105
    run_ID = "{}{}{}{}{}{}_{}".format(
106
        ("GMVAE" if variational else "AE"),
107
108
        ("Pred={}".format(predictor) if predictor > 0 and variational else ""),
        ("_Pheno={}".format(phenotype_class) if phenotype_class > 0 else ""),
109
        ("_loss={}".format(loss) if variational else ""),
110
111
        ("_encoding={}".format(logparam["encoding"]) if logparam is not None else ""),
        ("_k={}".format(logparam["k"]) if logparam is not None else ""),
112
        (datetime.now().strftime("%Y%m%d-%H%M%S")),
113
114
    )

115
    log_dir = os.path.abspath(os.path.join(outpath, "fit", run_ID))
116
    tensorboard_callback = tf.keras.callbacks.TensorBoard(
117
118
119
        log_dir=log_dir,
        histogram_freq=1,
        profile_batch=2,
120
121
122
    )

    onecycle = deepof.model_utils.one_cycle_scheduler(
123
124
        X_train.shape[0] // batch_size * 250,
        max_rate=0.005,
125
126
    )

127
128
129
130
    callbacks = [run_ID, tensorboard_callback, onecycle]

    if cp:
        cp_callback = tf.keras.callbacks.ModelCheckpoint(
131
            os.path.join(outpath, "checkpoints", run_ID + "/cp-{epoch:04d}.ckpt"),
132
133
134
135
136
137
138
139
            verbose=1,
            save_best_only=False,
            save_weights_only=True,
            save_freq="epoch",
        )
        callbacks.append(cp_callback)

    return callbacks
140
141


142
143
def deep_unsupervised_embedding(
    preprocessed_object: Tuple[np.ndarray, np.ndarray, np.ndarray, np.ndarray],
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
    batch_size: int,
    encoding_size: int,
    hparams: dict,
    kl_warmup: int,
    log_history: bool,
    log_hparams: bool,
    loss: str,
    mmd_warmup,
    montecarlo_kl,
    n_components,
    output_path,
    phenotype_class,
    predictor: float,
    pretrained: str,
    save_checkpoints: bool,
159
    save_weights: bool,
160
    variational: bool,
161
):
162
163
    """Implementation function for deepof.data.coordinates.deep_unsupervised_embedding"""

164
    # Load data
165
166
167
168
169
    X_train, y_train, X_val, y_val = preprocessed_object

    # To avoid stability issues
    tf.keras.backend.clear_session()

170
    # Defines what to log on tensorboard (useful for trying out different models)
171

172
173
    logparam = {
        "encoding": encoding_size,
174
        "k": n_components,
175
176
177
178
179
        "loss": loss,
    }
    if phenotype_class:
        logparam["pheno_weight"] = phenotype_class

180
    # Load callbacks
181
    run_ID, *cbacks = get_callbacks(
182
183
184
185
        X_train=X_train,
        batch_size=batch_size,
        cp=save_checkpoints,
        variational=variational,
186
        phenotype_class=phenotype_class,
187
188
189
190
191
        predictor=predictor,
        loss=loss,
        logparam=logparam,
        outpath=output_path,
    )
192
193
    if not log_history:
        cbacks = cbacks[1:]
194

195
    # Logs hyperparameters to tensorboard
196
197
    if log_hparams:
        logparams = [
198
            hp.HParam(
199
200
201
202
                "encoding",
                hp.Discrete([2, 4, 6, 8, 12, 16]),
                display_name="encoding",
                description="encoding size dimensionality",
203
            ),
204
205
206
207
208
209
210
211
212
213
214
            hp.HParam(
                "k",
                hp.IntInterval(min_value=1, max_value=25),
                display_name="k",
                description="cluster_number",
            ),
            hp.HParam(
                "loss",
                hp.Discrete(["ELBO", "MMD", "ELBO+MMD"]),
                display_name="loss function",
                description="loss function",
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
        rec = "reconstruction_" if phenotype_class else ""
        metrics = [
            hp.Metric("val_{}mae".format(rec), display_name="val_{}mae".format(rec)),
            hp.Metric("val_{}mse".format(rec), display_name="val_{}mse".format(rec)),
        ]
        if phenotype_class:
            logparams.append(
                hp.HParam(
                    "pheno_weight",
                    hp.RealInterval(min_value=0.0, max_value=1000.0),
                    display_name="pheno weight",
                    description="weight applied to phenotypic classifier from the latent space",
                )
            )
            metrics += [
                hp.Metric(
                    "phenotype_prediction_accuracy",
                    display_name="phenotype_prediction_accuracy",
                ),
                hp.Metric(
                    "phenotype_prediction_auc",
                    display_name="phenotype_prediction_auc",
                ),
            ]

        with tf.summary.create_file_writer(
            os.path.join(output_path, "hparams", run_ID)
        ).as_default():
            hp.hparams_config(
                hparams=logparams,
                metrics=metrics,
            )
250

251
    # Build models
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
    if not variational:
        encoder, decoder, ae = deepof.models.SEQ_2_SEQ_AE(
            ({} if hparams is None else hparams)
        ).build(X_train.shape)
        return_list = (encoder, decoder, ae)

    else:
        (
            encoder,
            generator,
            grouper,
            ae,
            kl_warmup_callback,
            mmd_warmup_callback,
        ) = deepof.models.SEQ_2_SEQ_GMVAE(
            architecture_hparams=({} if hparams is None else hparams),
            batch_size=batch_size,
            compile_model=True,
            encoding=encoding_size,
            kl_warmup_epochs=kl_warmup,
            loss=loss,
            mmd_warmup_epochs=mmd_warmup,
            montecarlo_kl=montecarlo_kl,
            neuron_control=False,
            number_of_components=n_components,
            overlap_loss=False,
            phenotype_prediction=phenotype_class,
            predictor=predictor,
        ).build(
            X_train.shape
        )
        return_list = (encoder, generator, grouper, ae)

    if pretrained:
286
        # If pretrained models are specified, load weights and return
287
288
289
290
291
292
293
294
295
296
297
298
299
        ae.load_weights(pretrained)
        return return_list

    else:
        if not variational:

            ae.fit(
                x=X_train,
                y=X_train,
                epochs=35,
                batch_size=batch_size,
                verbose=1,
                validation_data=(X_val, X_val),
300
301
                callbacks=cbacks
                + [
302
303
304
305
                    CustomStopper(
                        monitor="val_loss",
                        patience=5,
                        restore_best_weights=True,
306
                        start_epoch=max(kl_warmup, mmd_warmup),
307
308
309
310
311
312
                    ),
                ],
            )

        else:

313
            callbacks_ = cbacks + [
314
315
316
317
                CustomStopper(
                    monitor="val_loss",
                    patience=5,
                    restore_best_weights=True,
318
                    start_epoch=max(kl_warmup, mmd_warmup),
319
320
321
                ),
            ]

322
            if "ELBO" in loss and kl_warmup > 0:
323
324
                # noinspection PyUnboundLocalVariable
                callbacks_.append(kl_warmup_callback)
325
            if "MMD" in loss and mmd_warmup > 0:
326
327
328
                # noinspection PyUnboundLocalVariable
                callbacks_.append(mmd_warmup_callback)

329
330
331
332
333
334
335
            Xs, ys = [X_train], [X_train]
            Xvals, yvals = [X_val], [X_val]

            if predictor > 0.0:
                Xs, ys = X_train[:-1], [X_train[:-1], X_train[1:]]
                Xvals, yvals = X_val[:-1], [X_val[:-1], X_val[1:]]

336
            if phenotype_class > 0.0:
337
338
339
                ys += [y_train]
                yvals += [y_val]

340
            ae.fit(
341
342
343
344
345
346
347
348
349
350
351
352
                x=Xs,
                y=ys,
                epochs=35,
                batch_size=batch_size,
                verbose=1,
                validation_data=(
                    Xvals,
                    yvals,
                ),
                callbacks=callbacks_,
            )

353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
            if log_hparams:
                # noinspection PyUnboundLocalVariable
                def tensorboard_metric_logging(run_dir: str, hpms: Any):
                    output = gmvaep.predict(X_val)
                    if phenotype_class or predictor:
                        reconstruction = output[0]
                        prediction = output[1]
                        pheno = output[-1]
                    else:
                        reconstruction = output

                    with tf.summary.create_file_writer(run_dir).as_default():
                        hp.hparams(hpms)  # record the values used in this trial
                        val_mae = tf.reduce_mean(
                            tf.keras.metrics.mean_absolute_error(X_val, reconstruction)
368
                        )
369
370
                        val_mse = tf.reduce_mean(
                            tf.keras.metrics.mean_squared_error(X_val, reconstruction)
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
                        tf.summary.scalar("val_{}mae".format(rec), val_mae, step=1)
                        tf.summary.scalar("val_{}mse".format(rec), val_mse, step=1)

                        if predictor:
                            pred_mae = tf.reduce_mean(
                                tf.keras.metrics.mean_absolute_error(X_val, prediction)
                            )
                            pred_mse = tf.reduce_mean(
                                tf.keras.metrics.mean_squared_error(X_val, prediction)
                            )
                            tf.summary.scalar(
                                "val_prediction_mae".format(rec), pred_mae, step=1
                            )
                            tf.summary.scalar(
                                "val_prediction_mse".format(rec), pred_mse, step=1
                            )

                        if phenotype_class:
                            pheno_acc = tf.keras.metrics.binary_accuracy(
                                y_val, tf.squeeze(pheno)
                            )
                            pheno_auc = roc_auc_score(y_val, pheno)

                            tf.summary.scalar(
                                "phenotype_prediction_accuracy", pheno_acc, step=1
                            )
                            tf.summary.scalar(
                                "phenotype_prediction_auc", pheno_auc, step=1
                            )

                # Logparams to tensorboard
                tensorboard_metric_logging(
                    os.path.join(output_path, "hparams", run_ID),
                    logparam,
                )
407

408
409
410
    return return_list


411
def tune_search(
412
    data: List[np.array],
413
    encoding_size: int,
414
415
    hypertun_trials: int,
    hpt_type: str,
416
417
    hypermodel: str,
    k: int,
418
    kl_warmup_epochs: int,
419
    loss: str,
420
    mmd_warmup_epochs: int,
421
    overlap_loss: float,
422
    phenotype_class: float,
423
424
    predictor: float,
    project_name: str,
425
    callbacks: List,
426
    n_epochs: int = 30,
427
    n_replicas: int = 1,
428
) -> Union[bool, Tuple[Any, Any]]:
429
430
    """Define the search space using keras-tuner and bayesian optimization

431
432
433
434
435
436
437
438
439
440
441
    Parameters:
        - train (np.array): dataset to train the model on
        - test (np.array): dataset to validate the model on
        - hypertun_trials (int): number of Bayesian optimization iterations to run
        - hpt_type (str): specify one of Bayesian Optimization (bayopt) and Hyperband (hyperband)
        - hypermodel (str): hypermodel to load. Must be one of S2SAE (plain autoencoder)
        or S2SGMVAE (Gaussian Mixture Variational autoencoder).
        - k (int) number of components of the Gaussian Mixture
        - loss (str): one of [ELBO, MMD, ELBO+MMD]
        - overlap_loss (float): assigns as weight to an extra loss term which
        penalizes overlap between GM components
442
        - phenotype_class (float): adds an extra regularizing neural network to the model,
443
444
445
446
447
448
449
450
451
452
453
454
        which tries to predict the phenotype of the animal from which the sequence comes
        - predictor (float): adds an extra regularizing neural network to the model,
        which tries to predict the next frame from the current one
        - project_name (str): ID of the current run
        - callbacks (list): list of callbacks for the training loop
        - n_epochs (int): optional. Number of epochs to train each run for
        - n_replicas (int): optional. Number of replicas per parameter set. Higher values
         will yield more robust results, but will affect performance severely

    Returns:
        - best_hparams (dict): dictionary with the best retrieved hyperparameters
        - best_run (tf.keras.Model): trained instance of the best model found
455
456
457

    """

458
459
    X_train, y_train, X_val, y_val = data

460
461
462
463
    assert hpt_type in ["bayopt", "hyperband"], (
        "Invalid hyperparameter tuning framework. " "Select one of bayopt and hyperband"
    )

lucas_miranda's avatar
lucas_miranda committed
464
    if hypermodel == "S2SAE":  # pragma: no cover
465
        assert (
466
            predictor == 0.0 and phenotype_class == 0.0
467
        ), "Prediction branches are only available for variational models. See documentation for more details"
468
        hypermodel = deepof.hypermodels.SEQ_2_SEQ_AE(input_shape=X_train.shape)
469
470
471

    elif hypermodel == "S2SGMVAE":
        hypermodel = deepof.hypermodels.SEQ_2_SEQ_GMVAE(
472
            input_shape=X_train.shape,
473
            encoding=encoding_size,
474
            kl_warmup_epochs=kl_warmup_epochs,
475
            loss=loss,
476
            mmd_warmup_epochs=mmd_warmup_epochs,
477
            number_of_components=k,
478
            overlap_loss=overlap_loss,
479
            phenotype_predictor=phenotype_class,
480
            predictor=predictor,
481
        )
lucas_miranda's avatar
lucas_miranda committed
482

483
484
485
    else:
        return False

486
487
488
489
490
491
492
493
494
495
496
497
498
499
    hpt_params = {
        "hypermodel": hypermodel,
        "executions_per_trial": n_replicas,
        "logger": TensorBoardLogger(metrics=["val_mae"], logdir="./logs/hparams"),
        "objective": "val_mae",
        "project_name": project_name,
        "seed": 42,
        "tune_new_entries": True,
    }

    if hpt_type == "hyperband":
        tuner = Hyperband(
            directory="HyperBandx_{}_{}".format(loss, str(date.today())),
            max_epochs=hypertun_trials,
500
            hyperband_iterations=3,
lucas_miranda's avatar
lucas_miranda committed
501
            factor=2,
502
503
504
505
506
507
508
509
            **hpt_params
        )
    else:
        tuner = BayesianOptimization(
            directory="BayOpt_{}_{}".format(loss, str(date.today())),
            max_trials=hypertun_trials,
            **hpt_params
        )
510
511
512

    print(tuner.search_space_summary())

513
514
515
516
517
518
519
    Xs, ys = [X_train], [X_train]
    Xvals, yvals = [X_val], [X_val]

    if predictor > 0.0:
        Xs, ys = X_train[:-1], [X_train[:-1], X_train[1:]]
        Xvals, yvals = X_val[:-1], [X_val[:-1], X_val[1:]]

520
    if phenotype_class > 0.0:
521
522
523
        ys += [y_train]
        yvals += [y_val]

524
    tuner.search(
525
526
        Xs,
        ys,
527
        epochs=n_epochs,
528
        validation_data=(Xvals, yvals),
529
530
        verbose=1,
        batch_size=256,
lucas_miranda's avatar
lucas_miranda committed
531
        callbacks=callbacks,
532
533
534
535
536
    )

    best_hparams = tuner.get_best_hyperparameters(num_trials=1)[0]
    best_run = tuner.hypermodel.build(best_hparams)

lucas_miranda's avatar
lucas_miranda committed
537
538
    print(tuner.results_summary())

539
    return best_hparams, best_run