BrainMapperUNet.py 34.2 KB
Newer Older
1
2
3
4
"""Brain Mapper U-Net Architecture

Description:

5
6
    This folder contains the Pytorch implementation of the core U-net architecture.
    This arcitecture predicts functional connectivity rsfMRI from structural connectivity information from dMRI.
7

8
9
10
11
12
13
Usage:

    To use this module, import it and instantiate is as you wish:

        from BrainMapperUNet import BrainMapperUNet
        deep_learning_model = BrainMapperUnet(parameters)
14
15
16
17
18
19

"""

import numpy as np
import torch
import torch.nn as nn
20
import utils.modules as modules
21

22

23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
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
class BrainMapperResUNet3D(nn.Module):
    """Architecture class for Residual DenseBlock BrainMapper 3D U-net.

    This class contains the pytorch implementation of the U-net architecture underpinning the BrainMapper project.

    Args:
        parameters (dict): Contains information relevant parameters
        parameters = {
            'kernel_heigth': 5
            'kernel_width': 5
            'kernel_depth': 5
            'kernel_classification': 1
            'input_channels': 1
            'output_channels': 64
            'convolution_stride': 1
            'dropout': 0.2
            'pool_kernel_size': 2
            'pool_stride': 2
            'up_mode': 'upconv'
            'number_of_classes': 1
        }

    Returns:
        probability_map (torch.tensor): Output forward passed tensor through the U-net block
    """

    def __init__(self, parameters):
        super(BrainMapperResUNet3D, self).__init__()

        original_input_channels = parameters['input_channels']
        original_output_channels = parameters['output_channels']

        self.encoderBlock1 = modules.DensEncoderBlock3D(parameters)
        parameters['input_channels'] = parameters['output_channels']
        self.encoderBlock2 = modules.DensEncoderBlock3D(parameters)
        self.encoderBlock3 = modules.DensEncoderBlock3D(parameters)
        self.encoderBlock4 = modules.DensEncoderBlock3D(parameters)

        self.bottleneck = modules.DensBlock3D(parameters)

        parameters['input_channels'] = parameters['output_channels'] * 2
        self.decoderBlock1 = modules.DensDecoderBlock3D(parameters)
        self.decoderBlock2 = modules.DensDecoderBlock3D(parameters)
        self.decoderBlock3 = modules.DensDecoderBlock3D(parameters)
        self.decoderBlock4 = modules.DensDecoderBlock3D(parameters)

        parameters['input_channels'] = parameters['output_channels']
        self.classifier = modules.DensClassifierBlock3D(parameters)

        parameters['input_channels'] = original_input_channels
        parameters['output_channels'] = original_output_channels

    def forward(self, X):
        """Forward pass for 3D U-net

        Function computing the forward pass through the 3D U-Net
        The input to the function is the dMRI map

        Args:
            X (torch.tensor): Input dMRI map, shape = (N x C x D x H x W) 

        Returns:
            probability_map (torch.tensor): Output forward passed tensor through the U-net block
        """

        Y_encoder_1, Y_np1, _ = self.encoderBlock1.forward(X)
        Y_encoder_2, Y_np2, _ = self.encoderBlock2.forward(
            Y_encoder_1)

        del Y_encoder_1

        Y_encoder_3, Y_np3, _ = self.encoderBlock3.forward(
            Y_encoder_2)

        del Y_encoder_2

        Y_encoder_4, Y_np4, _ = self.encoderBlock4.forward(
            Y_encoder_3)

        del Y_encoder_3

        Y_bottleNeck = self.bottleneck.forward(Y_encoder_4)

        del Y_encoder_4

        Y_decoder_1 = self.decoderBlock1.forward(
            Y_bottleNeck, Y_np4)

        del Y_bottleNeck, Y_np4

        Y_decoder_2 = self.decoderBlock2.forward(
            Y_decoder_1, Y_np3)

        del Y_decoder_1, Y_np3

        Y_decoder_3 = self.decoderBlock3.forward(
            Y_decoder_2, Y_np2)

        del Y_decoder_2, Y_np2

        Y_decoder_4 = self.decoderBlock4.forward(
            Y_decoder_3, Y_np1)

        del Y_decoder_3, Y_np1

        probability_map = self.classifier.forward(Y_decoder_4)

        del Y_decoder_4

        return probability_map

    def save(self, path):
        """Model Saver

        Function saving the model with all its parameters to a given path.
        The path must end with a *.model argument.

        Args:
            path (str): Path string
        """

        print("Saving Model... {}".format(path))
        torch.save(self, path)

    @property
    def test_if_cuda(self):
        """Cuda Test

        This function tests if the model parameters are allocated to a CUDA enabled GPU.

        Returns:
            bool: Flag indicating True if the tensor is stored on the GPU and Flase otherwhise
        """

        return next(self.parameters()).is_cuda

    def predict(self, X, device=0):
        """Post-training Output Prediction

        This function predicts the output of the of the U-net post-training

        Args:
            X (torch.tensor): input dMRI volume
            device (int/str): Device type used for training (int - GPU id, str- CPU)

        Returns:
            prediction (ndarray): predicted output after training

        """
        self.eval()  # PyToch module setting network to evaluation mode

        if type(X) is np.ndarray:
            X = torch.tensor(X, requires_grad=False).type(torch.FloatTensor)
        elif type(X) is torch.Tensor and not X.is_cuda:
            X = X.type(torch.FloatTensor).cuda(device, non_blocking=True)

        # .cuda() call transfers the densor from the CPU to the GPU if that is the case.
        # Non-blocking argument lets the caller bypas synchronization when necessary

        with torch.no_grad():  # Causes operations to have no gradients
            output = self.forward(X)

        _, idx = torch.max(output, 1)

        # We retrieve the tensor held by idx (.data), and map it to a cpu as an ndarray
        idx = idx.data.cpu().numpy()

        prediction = np.squeeze(idx)

        del X, output, idx

        return prediction

    def reset_parameters(self):
        """Parameter Initialization

        This function (re)initializes the parameters of the defined network.
        This function is a wrapper for the reset_parameters() function defined for each module. 
        More information can be found here: https://discuss.pytorch.org/t/what-is-the-default-initialization-of-a-conv2d-layer-and-linear-layer/16055 + https://discuss.pytorch.org/t/how-to-reset-model-weights-to-effectively-implement-crossvalidation/53859 
        An alternative (re)initialization method is described here: https://discuss.pytorch.org/t/how-to-reset-variables-values-in-nn-modules/32639 
        """

        print("Initializing network parameters...")

        for _, module in self.named_children():
            for _, submodule in module.named_children():
                for _, subsubmodule in submodule.named_children():
                    if isinstance(subsubmodule, (torch.nn.PReLU, torch.nn.Dropout3d, torch.nn.MaxPool3d)) == False:
                        subsubmodule.reset_parameters()

        print("Initialized network parameters!")


216
class BrainMapperUNet3D(nn.Module):
217
    """Architecture class for Traditional BrainMapper 3D U-net.
218
219
220
221
222
223
224
225

    This class contains the pytorch implementation of the U-net architecture underpinning the BrainMapper project.

    Args:
        parameters (dict): Contains information relevant parameters
        parameters = {
            'kernel_heigth': 5
            'kernel_width': 5
226
            'kernel_depth': 5
227
228
229
230
231
232
233
234
235
236
            'kernel_classification': 1
            'input_channels': 1
            'output_channels': 64
            'convolution_stride': 1
            'dropout': 0.2
            'pool_kernel_size': 2
            'pool_stride': 2
            'up_mode': 'upconv'
            'number_of_classes': 1
        }
237

238
239
    Returns:
        probability_map (torch.tensor): Output forward passed tensor through the U-net block
240
    """
241

242
    def __init__(self, parameters):
243
        super(BrainMapperUNet3D, self).__init__()
244

245
246
        original_input_channels = parameters['input_channels']
        original_output_channels = parameters['output_channels']
247

248
        self.encoderBlock1 = modules.EncoderBlock3D(parameters)
249
        parameters['input_channels'] = parameters['output_channels']
250
251
252
253
254
255
256
257
        parameters['output_channels'] = parameters['output_channels'] * 2
        self.encoderBlock2 = modules.EncoderBlock3D(parameters)
        parameters['input_channels'] = parameters['output_channels']
        parameters['output_channels'] = parameters['output_channels'] * 2
        self.encoderBlock3 = modules.EncoderBlock3D(parameters)
        parameters['input_channels'] = parameters['output_channels']
        parameters['output_channels'] = parameters['output_channels'] * 2
        self.encoderBlock4 = modules.EncoderBlock3D(parameters)
258

259
260
261
        parameters['input_channels'] = parameters['output_channels']
        parameters['output_channels'] = parameters['output_channels'] * 2
        self.bottleneck = modules.ConvolutionalBlock3D(parameters)
262

263
264
265
266
267
268
269
270
271
272
273
274
        parameters['input_channels'] = parameters['output_channels']
        parameters['output_channels'] = parameters['output_channels'] // 2
        self.decoderBlock1 = modules.DecoderBlock3D(parameters)
        parameters['input_channels'] = parameters['output_channels']
        parameters['output_channels'] = parameters['output_channels'] // 2
        self.decoderBlock2 = modules.DecoderBlock3D(parameters)
        parameters['input_channels'] = parameters['output_channels']
        parameters['output_channels'] = parameters['output_channels'] // 2
        self.decoderBlock3 = modules.DecoderBlock3D(parameters)
        parameters['input_channels'] = parameters['output_channels']
        parameters['output_channels'] = parameters['output_channels'] // 2
        self.decoderBlock4 = modules.DecoderBlock3D(parameters)
275
276

        parameters['input_channels'] = parameters['output_channels']
277
278
279
280
        self.classifier = modules.ClassifierBlock3D(parameters)

        parameters['input_channels'] = original_input_channels
        parameters['output_channels'] = original_output_channels
281
282

    def forward(self, X):
283
        """Forward pass for 3D U-net
284

285
        Function computing the forward pass through the 3D U-Net
286
287
288
        The input to the function is the dMRI map

        Args:
289
            X (torch.tensor): Input dMRI map, shape = (N x C x D x H x W) 
290
291
292

        Returns:
            probability_map (torch.tensor): Output forward passed tensor through the U-net block
293
        """
294

295
296
        Y_encoder_1, Y_np1, _ = self.encoderBlock1.forward(X)
        Y_encoder_2, Y_np2, _ = self.encoderBlock2.forward(
297
            Y_encoder_1)
298
299
300

        del Y_encoder_1

301
        Y_encoder_3, Y_np3, _ = self.encoderBlock3.forward(
302
            Y_encoder_2)
303
304
305

        del Y_encoder_2

306
        Y_encoder_4, Y_np4, _ = self.encoderBlock4.forward(
307
308
            Y_encoder_3)

309
310
        del Y_encoder_3

311
312
        Y_bottleNeck = self.bottleneck.forward(Y_encoder_4)

313
314
        del Y_encoder_4

315
        Y_decoder_1 = self.decoderBlock1.forward(
316
            Y_bottleNeck, Y_np4)
317

318
        del Y_bottleNeck, Y_np4
319

320
        Y_decoder_2 = self.decoderBlock2.forward(
321
            Y_decoder_1, Y_np3)
322

323
        del Y_decoder_1, Y_np3
324

325
        Y_decoder_3 = self.decoderBlock3.forward(
326
            Y_decoder_2, Y_np2)
327

328
        del Y_decoder_2, Y_np2
329

330
        Y_decoder_4 = self.decoderBlock4.forward(
331
            Y_decoder_3, Y_np1)
332

333
        del Y_decoder_3, Y_np1
334

335
336
        probability_map = self.classifier.forward(Y_decoder_4)

337
338
        del Y_decoder_4

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
        return probability_map

    def save(self, path):
        """Model Saver

        Function saving the model with all its parameters to a given path.
        The path must end with a *.model argument.

        Args:
            path (str): Path string
        """

        print("Saving Model... {}".format(path))
        torch.save(self, path)

    @property
    def test_if_cuda(self):
        """Cuda Test

        This function tests if the model parameters are allocated to a CUDA enabled GPU.

        Returns:
            bool: Flag indicating True if the tensor is stored on the GPU and Flase otherwhise
        """

        return next(self.parameters()).is_cuda

    def predict(self, X, device=0):
        """Post-training Output Prediction

        This function predicts the output of the of the U-net post-training

        Args:
            X (torch.tensor): input dMRI volume
            device (int/str): Device type used for training (int - GPU id, str- CPU)

        Returns:
            prediction (ndarray): predicted output after training

        """
        self.eval()  # PyToch module setting network to evaluation mode

        if type(X) is np.ndarray:
            X = torch.tensor(X, requires_grad=False).type(torch.FloatTensor)
        elif type(X) is torch.Tensor and not X.is_cuda:
            X = X.type(torch.FloatTensor).cuda(device, non_blocking=True)

        # .cuda() call transfers the densor from the CPU to the GPU if that is the case.
        # Non-blocking argument lets the caller bypas synchronization when necessary

        with torch.no_grad():  # Causes operations to have no gradients
            output = self.forward(X)

        _, idx = torch.max(output, 1)

        # We retrieve the tensor held by idx (.data), and map it to a cpu as an ndarray
        idx = idx.data.cpu().numpy()

        prediction = np.squeeze(idx)

        del X, output, idx

        return prediction

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
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
    def reset_parameters(self):
        """Parameter Initialization

        This function (re)initializes the parameters of the defined network.
        This function is a wrapper for the reset_parameters() function defined for each module. 
        More information can be found here: https://discuss.pytorch.org/t/what-is-the-default-initialization-of-a-conv2d-layer-and-linear-layer/16055 + https://discuss.pytorch.org/t/how-to-reset-model-weights-to-effectively-implement-crossvalidation/53859 
        An alternative (re)initialization method is described here: https://discuss.pytorch.org/t/how-to-reset-variables-values-in-nn-modules/32639 
        """

        print("Initializing network parameters...")

        for _, module in self.named_children():
            for _, submodule in module.named_children():
                for _, subsubmodule in submodule.named_children():
                    if isinstance(subsubmodule, (torch.nn.PReLU, torch.nn.Dropout3d, torch.nn.MaxPool3d)) == False:
                        subsubmodule.reset_parameters()

        print("Initialized network parameters!")


class BrainMapperUNet3Dsimple(nn.Module):
    """Architecture class for  Simple BrainMapper 3D U-net.

    This class contains the pytorch implementation of the U-net architecture underpinning the BrainMapper project.

    Args:
        parameters (dict): Contains information relevant parameters
        parameters = {
            'kernel_heigth': 5
            'kernel_width': 5
            'kernel_depth': 5
            'kernel_classification': 1
            'input_channels': 1
            'output_channels': 64
            'convolution_stride': 1
            'dropout': 0.2
            'pool_kernel_size': 2
            'pool_stride': 2
            'up_mode': 'upconv'
            'number_of_classes': 1
        }

    Returns:
        probability_map (torch.tensor): Output forward passed tensor through the U-net block
    """

    def __init__(self, parameters):
        super(BrainMapperUNet3D, self).__init__()

        original_input_channels = parameters['input_channels']
        original_output_channels = parameters['output_channels']

        self.encoderBlock1 = modules.EncoderBlock3Dsimple(parameters)
        parameters['input_channels'] = parameters['output_channels']
        parameters['output_channels'] = parameters['output_channels'] * 2
        self.encoderBlock2 = modules.EncoderBlock3Dsimple(parameters)
        parameters['input_channels'] = parameters['output_channels']
        parameters['output_channels'] = parameters['output_channels'] * 2
        self.encoderBlock3 = modules.EncoderBlock3Dsimple(parameters)
        parameters['input_channels'] = parameters['output_channels']
        parameters['output_channels'] = parameters['output_channels'] * 2
        self.encoderBlock4 = modules.EncoderBlock3Dsimple(parameters)

        parameters['input_channels'] = parameters['output_channels']
        parameters['output_channels'] = parameters['output_channels'] * 2
        self.bottleneck = modules.ConvolutionalBlock3Dsimple(parameters)

        parameters['input_channels'] = parameters['output_channels']
        parameters['output_channels'] = parameters['output_channels'] // 2
        self.decoderBlock1 = modules.DecoderBlock3Dsimple(parameters)
        parameters['input_channels'] = parameters['output_channels']
        parameters['output_channels'] = parameters['output_channels'] // 2
        self.decoderBlock2 = modules.DecoderBlock3Dsimple(parameters)
        parameters['input_channels'] = parameters['output_channels']
        parameters['output_channels'] = parameters['output_channels'] // 2
        self.decoderBlock3 = modules.DecoderBlock3Dsimple(parameters)
        parameters['input_channels'] = parameters['output_channels']
        parameters['output_channels'] = parameters['output_channels'] // 2
        self.decoderBlock4 = modules.DecoderBlock3Dsimple(parameters)

        parameters['input_channels'] = parameters['output_channels']
        self.classifier = modules.ClassifierBlock3Dsimple(parameters)

        parameters['input_channels'] = original_input_channels
        parameters['output_channels'] = original_output_channels

    def forward(self, X):
        """Forward pass for 3D U-net

        Function computing the forward pass through the 3D U-Net
        The input to the function is the dMRI map

        Args:
            X (torch.tensor): Input dMRI map, shape = (N x C x D x H x W) 

        Returns:
            probability_map (torch.tensor): Output forward passed tensor through the U-net block
        """

        Y_encoder_1, Y_np1, _ = self.encoderBlock1.forward(X)
        Y_encoder_2, Y_np2, _ = self.encoderBlock2.forward(
            Y_encoder_1)

        del Y_encoder_1

        Y_encoder_3, Y_np3, _ = self.encoderBlock3.forward(
            Y_encoder_2)

        del Y_encoder_2

        Y_encoder_4, Y_np4, _ = self.encoderBlock4.forward(
            Y_encoder_3)

        del Y_encoder_3

        Y_bottleNeck = self.bottleneck.forward(Y_encoder_4)

        del Y_encoder_4

        Y_decoder_1 = self.decoderBlock1.forward(
            Y_bottleNeck, Y_np4)

        del Y_bottleNeck, Y_np4

        Y_decoder_2 = self.decoderBlock2.forward(
            Y_decoder_1, Y_np3)

        del Y_decoder_1, Y_np3

        Y_decoder_3 = self.decoderBlock3.forward(
            Y_decoder_2, Y_np2)

        del Y_decoder_2, Y_np2

        Y_decoder_4 = self.decoderBlock4.forward(
            Y_decoder_3, Y_np1)

        del Y_decoder_3, Y_np1

        probability_map = self.classifier.forward(Y_decoder_4)

        del Y_decoder_4

        return probability_map

    def save(self, path):
        """Model Saver

        Function saving the model with all its parameters to a given path.
        The path must end with a *.model argument.

        Args:
            path (str): Path string
        """

        print("Saving Model... {}".format(path))
        torch.save(self, path)

    @property
    def test_if_cuda(self):
        """Cuda Test

        This function tests if the model parameters are allocated to a CUDA enabled GPU.

        Returns:
            bool: Flag indicating True if the tensor is stored on the GPU and Flase otherwhise
        """

        return next(self.parameters()).is_cuda

    def predict(self, X, device=0):
        """Post-training Output Prediction

        This function predicts the output of the of the U-net post-training

        Args:
            X (torch.tensor): input dMRI volume
            device (int/str): Device type used for training (int - GPU id, str- CPU)

        Returns:
            prediction (ndarray): predicted output after training

        """
        self.eval()  # PyToch module setting network to evaluation mode

        if type(X) is np.ndarray:
            X = torch.tensor(X, requires_grad=False).type(torch.FloatTensor)
        elif type(X) is torch.Tensor and not X.is_cuda:
            X = X.type(torch.FloatTensor).cuda(device, non_blocking=True)

        # .cuda() call transfers the densor from the CPU to the GPU if that is the case.
        # Non-blocking argument lets the caller bypas synchronization when necessary

        with torch.no_grad():  # Causes operations to have no gradients
            output = self.forward(X)

        _, idx = torch.max(output, 1)

        # We retrieve the tensor held by idx (.data), and map it to a cpu as an ndarray
        idx = idx.data.cpu().numpy()

        prediction = np.squeeze(idx)

        del X, output, idx

        return prediction

610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
    def reset_parameters(self):
        """Parameter Initialization

        This function (re)initializes the parameters of the defined network.
        This function is a wrapper for the reset_parameters() function defined for each module. 
        More information can be found here: https://discuss.pytorch.org/t/what-is-the-default-initialization-of-a-conv2d-layer-and-linear-layer/16055 + https://discuss.pytorch.org/t/how-to-reset-model-weights-to-effectively-implement-crossvalidation/53859 
        An alternative (re)initialization method is described here: https://discuss.pytorch.org/t/how-to-reset-variables-values-in-nn-modules/32639 
        """

        print("Initializing network parameters...")

        for _, module in self.named_children():
            for _, submodule in module.named_children():
                if isinstance(submodule, (torch.nn.PReLU, torch.nn.Dropout3d, torch.nn.MaxPool3d)) == False:
                    submodule.reset_parameters()

        print("Initialized network parameters!")

628

629
630
# DEPRECATED ARCHITECTURES!

631

632
633
class BrainMapperUNet(nn.Module):
    """Architecture class BrainMapper U-net.
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656

    This class contains the pytorch implementation of the U-net architecture underpinning the BrainMapper project.

    Args:
        parameters (dict): Contains information relevant parameters
        parameters = {
            'kernel_heigth': 5
            'kernel_width': 5
            'kernel_classification': 1
            'input_channels': 1
            'output_channels': 64
            'convolution_stride': 1
            'dropout': 0.2
            'pool_kernel_size': 2
            'pool_stride': 2
            'up_mode': 'upconv'
            'number_of_classes': 1
        }

    Returns:
        probability_map (torch.tensor): Output forward passed tensor through the U-net block
    """

657
    def __init__(self, parameters):
658
        super(BrainMapperUNet, self).__init__()
659

660
        # TODO: currently, architecture based on QuickNAT - need to adjust parameter values accordingly!
661

662
        self.encoderBlock1 = modules.EncoderBlock(parameters)
663
        parameters['input_channels'] = parameters['output_channels']
664
665
666
        self.encoderBlock2 = modules.EncoderBlock(parameters)
        self.encoderBlock3 = modules.EncoderBlock(parameters)
        self.encoderBlock4 = modules.EncoderBlock(parameters)
667

668
        self.bottleneck = modules.ConvolutionalBlock(parameters)
669

670
671
672
673
674
        parameters['input_channels'] = parameters['output_channels'] * 2.0
        self.decoderBlock1 = modules.DecoderBlock(parameters)
        self.decoderBlock2 = modules.DecoderBlock(parameters)
        self.decoderBlock3 = modules.DecoderBlock(parameters)
        self.decoderBlock4 = modules.DecoderBlock(parameters)
675
676

        parameters['input_channels'] = parameters['output_channels']
677
        self.classifier = modules.ClassifierBlock(parameters)
678
679

    def forward(self, X):
680
        """Forward pass for U-net
681

682
        Function computing the forward pass through the U-Net
683
684
685
        The input to the function is the dMRI map

        Args:
686
            X (torch.tensor): Input dMRI map, shape = (N x C x H x W) 
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

        Returns:
            probability_map (torch.tensor): Output forward passed tensor through the U-net block
        """

        Y_encoder_1, Y_np1, pool_indices1 = self.encoderBlock1.forward(X)
        Y_encoder_2, Y_np2, pool_indices2 = self.encoderBlock2.forward(
            Y_encoder_1)

        del Y_encoder_1

        Y_encoder_3, Y_np3, pool_indices3 = self.encoderBlock3.forward(
            Y_encoder_2)

        del Y_encoder_2

        Y_encoder_4, Y_np4, pool_indices4 = self.encoderBlock4.forward(
            Y_encoder_3)

        del Y_encoder_3

        Y_bottleNeck = self.bottleneck.forward(Y_encoder_4)

        del Y_encoder_4

        Y_decoder_1 = self.decoderBlock1.forward(
            Y_bottleNeck, Y_np4, pool_indices4)

        del Y_bottleNeck, Y_np4, pool_indices4
716

717
718
719
720
721
722
723
724
725
726
        Y_decoder_2 = self.decoderBlock2.forward(
            Y_decoder_1, Y_np3, pool_indices3)

        del Y_decoder_1, Y_np3, pool_indices3

        Y_decoder_3 = self.decoderBlock3.forward(
            Y_decoder_2, Y_np2, pool_indices2)

        del Y_decoder_2, Y_np2, pool_indices2

727
        Y_decoder_4 = self.decoderBlock4.forward(
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
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
            Y_decoder_3, Y_np1, pool_indices1)

        del Y_decoder_3, Y_np1, pool_indices1

        probability_map = self.classifier.forward(Y_decoder_4)

        del Y_decoder_4

        return probability_map

    def save(self, path):
        """Model Saver

        Function saving the model with all its parameters to a given path.
        The path must end with a *.model argument.

        Args:
            path (str): Path string
        """

        print("Saving Model... {}".format(path))
        torch.save(self, path)

    @property
    def test_if_cuda(self):
        """Cuda Test

        This function tests if the model parameters are allocated to a CUDA enabled GPU.

        Returns:
            bool: Flag indicating True if the tensor is stored on the GPU and Flase otherwhise
        """

        return next(self.parameters()).is_cuda

    def predict(self, X, device=0):
        """Post-training Output Prediction

        This function predicts the output of the of the U-net post-training

        Args:
            X (torch.tensor): input dMRI volume
            device (int/str): Device type used for training (int - GPU id, str- CPU)

        Returns:
            prediction (ndarray): predicted output after training

        """
        self.eval()  # PyToch module setting network to evaluation mode

        if type(X) is np.ndarray:
            X = torch.tensor(X, requires_grad=False).type(torch.FloatTensor)
        elif type(X) is torch.Tensor and not X.is_cuda:
            X = X.type(torch.FloatTensor).cuda(device, non_blocking=True)

        # .cuda() call transfers the densor from the CPU to the GPU if that is the case.
        # Non-blocking argument lets the caller bypas synchronization when necessary

        with torch.no_grad():  # Causes operations to have no gradients
            output = self.forward(X)

        _, idx = torch.max(output, 1)

        # We retrieve the tensor held by idx (.data), and map it to a cpu as an ndarray
        idx = idx.data.cpu().numpy()

        prediction = np.squeeze(idx)

        del X, output, idx

        return prediction

800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
    def reset_parameters(self):
        """Parameter Initialization

        This function (re)initializes the parameters of the defined network.
        This function is a wrapper for the reset_parameters() function defined for each module. 
        More information can be found here: https://discuss.pytorch.org/t/what-is-the-default-initialization-of-a-conv2d-layer-and-linear-layer/16055 + https://discuss.pytorch.org/t/how-to-reset-model-weights-to-effectively-implement-crossvalidation/53859 
        An alternative (re)initialization method is described here: https://discuss.pytorch.org/t/how-to-reset-variables-values-in-nn-modules/32639 
        """

        print("Initializing network parameters...")

        for _, module in self.named_children():
            for _, submodule in module.named_children():
                if isinstance(submodule, (torch.nn.PReLU, torch.nn.Dropout3d, torch.nn.MaxPool3d)) == False:
                    submodule.reset_parameters()

        print("Initialized network parameters!")
817

818

819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
class BrainMapperUNet3D_Simple(nn.Module):
    """Architecture class BrainMapper 3D U-net.

    This class contains the pytorch implementation of the U-net architecture underpinning the BrainMapper project.

    Args:
        parameters (dict): Contains information relevant parameters
        parameters = {
            'kernel_heigth': 5
            'kernel_width': 5
            'kernel_depth': 5
            'kernel_classification': 1
            'input_channels': 1
            'output_channels': 64
            'convolution_stride': 1
            'dropout': 0.2
            'pool_kernel_size': 2
            'pool_stride': 2
            'up_mode': 'upconv'
            'number_of_classes': 1
        }

    Returns:
        probability_map (torch.tensor): Output forward passed tensor through the U-net block
    """

845
    def __init__(self, parameters):
846
        super(BrainMapperUNet3D_Simple, self).__init__()
847
848
849
850
851
852
853
854
855
856
857

        # TODO: currently, architecture based on QuickNAT - need to adjust parameter values accordingly!

        self.encoderBlock1 = modules.EncoderBlock3D(parameters)
        parameters['input_channels'] = parameters['output_channels']
        self.encoderBlock2 = modules.EncoderBlock3D(parameters)
        self.encoderBlock3 = modules.EncoderBlock3D(parameters)
        self.encoderBlock4 = modules.EncoderBlock3D(parameters)

        self.bottleneck = modules.ConvolutionalBlock3D(parameters)

858
        parameters['input_channels'] = parameters['output_channels'] * 2
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
        self.decoderBlock1 = modules.DecoderBlock3D(parameters)
        self.decoderBlock2 = modules.DecoderBlock3D(parameters)
        self.decoderBlock3 = modules.DecoderBlock3D(parameters)
        self.decoderBlock4 = modules.DecoderBlock3D(parameters)

        parameters['input_channels'] = parameters['output_channels']
        self.classifier = modules.ClassifierBlock3D(parameters)

    def forward(self, X):
        """Forward pass for 3D U-net

        Function computing the forward pass through the 3D U-Net
        The input to the function is the dMRI map

        Args:
            X (torch.tensor): Input dMRI map, shape = (N x C x D x H x W) 

        Returns:
            probability_map (torch.tensor): Output forward passed tensor through the U-net block
        """

880
        Y_encoder_1, Y_np1, pool_indices1 = self.encoderBlock1.forward(X)
881
882
        Y_encoder_2, Y_np2, pool_indices2 = self.encoderBlock2.forward(
            Y_encoder_1)
883
884
885

        del Y_encoder_1

886
887
        Y_encoder_3, Y_np3, pool_indices3 = self.encoderBlock3.forward(
            Y_encoder_2)
888
889
890

        del Y_encoder_2

891
892
        Y_encoder_4, Y_np4, pool_indices4 = self.encoderBlock4.forward(
            Y_encoder_3)
893

894
895
        del Y_encoder_3

896
897
        Y_bottleNeck = self.bottleneck.forward(Y_encoder_4)

898
899
        del Y_encoder_4

900
901
        Y_decoder_1 = self.decoderBlock1.forward(
            Y_bottleNeck, Y_np4, pool_indices4)
902
903

        del Y_bottleNeck, Y_np4, pool_indices4
904

905
906
        Y_decoder_2 = self.decoderBlock2.forward(
            Y_decoder_1, Y_np3, pool_indices3)
907
908
909

        del Y_decoder_1, Y_np3, pool_indices3

910
911
        Y_decoder_3 = self.decoderBlock3.forward(
            Y_decoder_2, Y_np2, pool_indices2)
912
913
914

        del Y_decoder_2, Y_np2, pool_indices2

915
        Y_decoder_4 = self.decoderBlock4.forward(
916
            Y_decoder_3, Y_np1, pool_indices1)
917

918
919
920
921
922
923
        del Y_decoder_3, Y_np1, pool_indices1

        probability_map = self.classifier.forward(Y_decoder_4)

        del Y_decoder_4

924
        probability_map = self.classifier.forward(Y_decoder_4)
925

926
        return probability_map
927

928
    def save(self, path):
Andrei-Claudiu Roibu's avatar
Andrei-Claudiu Roibu committed
929
        """Model Saver
930

Andrei-Claudiu Roibu's avatar
Andrei-Claudiu Roibu committed
931
932
        Function saving the model with all its parameters to a given path.
        The path must end with a *.model argument.
933

Andrei-Claudiu Roibu's avatar
Andrei-Claudiu Roibu committed
934
935
936
        Args:
            path (str): Path string
        """
937

Andrei-Claudiu Roibu's avatar
Andrei-Claudiu Roibu committed
938
939
        print("Saving Model... {}".format(path))
        torch.save(self, path)
940
941
942
943
944
945
946
947
948
949
950
951

    @property
    def test_if_cuda(self):
        """Cuda Test

        This function tests if the model parameters are allocated to a CUDA enabled GPU.

        Returns:
            bool: Flag indicating True if the tensor is stored on the GPU and Flase otherwhise
        """

        return next(self.parameters()).is_cuda
952
953

    def predict(self, X, device=0):
954
        """Post-training Output Prediction
955

956
957
958
959
        This function predicts the output of the of the U-net post-training

        Args:
            X (torch.tensor): input dMRI volume
960
            device (int/str): Device type used for training (int - GPU id, str- CPU)
961
962
963
964

        Returns:
            prediction (ndarray): predicted output after training

965
        """
966
        self.eval()  # PyToch module setting network to evaluation mode
967
968
969
970
971
972
973
974
975

        if type(X) is np.ndarray:
            X = torch.tensor(X, requires_grad=False).type(torch.FloatTensor)
        elif type(X) is torch.Tensor and not X.is_cuda:
            X = X.type(torch.FloatTensor).cuda(device, non_blocking=True)

        # .cuda() call transfers the densor from the CPU to the GPU if that is the case.
        # Non-blocking argument lets the caller bypas synchronization when necessary

976
        with torch.no_grad():  # Causes operations to have no gradients
977
978
979
980
            output = self.forward(X)

        _, idx = torch.max(output, 1)

981
982
983
        # We retrieve the tensor held by idx (.data), and map it to a cpu as an ndarray
        idx = idx.data.cpu().numpy()

984
985
986
987
988
989
        prediction = np.squeeze(idx)

        del X, output, idx

        return prediction

990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
    def reset_parameters(self):
        """Parameter Initialization

        This function (re)initializes the parameters of the defined network.
        This function is a wrapper for the reset_parameters() function defined for each module. 
        More information can be found here: https://discuss.pytorch.org/t/what-is-the-default-initialization-of-a-conv2d-layer-and-linear-layer/16055 + https://discuss.pytorch.org/t/how-to-reset-model-weights-to-effectively-implement-crossvalidation/53859 
        An alternative (re)initialization method is described here: https://discuss.pytorch.org/t/how-to-reset-variables-values-in-nn-modules/32639 
        """

        print("Initializing network parameters...")

        for _, module in self.named_children():
            for _, submodule in module.named_children():
                if isinstance(submodule, (torch.nn.PReLU, torch.nn.Dropout3d, torch.nn.MaxPool3d)) == False:
                    submodule.reset_parameters()

        print("Initialized network parameters!")

1008