CNNSupervisedTrainer_VGG16.py 17.7 KB
Newer Older
Nicola Gatto's avatar
Nicola Gatto committed
1
2
3
4
5
6
import mxnet as mx
import logging
import numpy as np
import time
import os
import shutil
Christian Fuß's avatar
Christian Fuß committed
7
import pickle
Sebastian Nickels's avatar
Sebastian Nickels committed
8
9
import math
import sys
Nicola Gatto's avatar
Nicola Gatto committed
10
11
from mxnet import gluon, autograd, nd

Eyüp Harputlu's avatar
Eyüp Harputlu committed
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
class CrossEntropyLoss(gluon.loss.Loss):
    def __init__(self, axis=-1, sparse_label=True, weight=None, batch_axis=0, **kwargs):
        super(CrossEntropyLoss, self).__init__(weight, batch_axis, **kwargs)
        self._axis = axis
        self._sparse_label = sparse_label

    def hybrid_forward(self, F, pred, label, sample_weight=None):
        pred = F.log(pred)
        if self._sparse_label:
            loss = -F.pick(pred, label, axis=self._axis, keepdims=True)
        else:
            label = gluon.loss._reshape_like(F, label, pred)
            loss = -F.sum(pred * label, axis=self._axis, keepdims=True)
        loss = gluon.loss._apply_weighting(F, loss, self._weight, sample_weight)
        return F.mean(loss, axis=self._batch_axis, exclude=True)

Eyüp Harputlu's avatar
Eyüp Harputlu committed
28
29
30
31
32
33
34
35
36
class LogCoshLoss(gluon.loss.Loss):
    def __init__(self, weight=None, batch_axis=0, **kwargs):
        super(LogCoshLoss, self).__init__(weight, batch_axis, **kwargs)

    def hybrid_forward(self, F, pred, label, sample_weight=None):
        loss = F.log(F.cosh(pred - label))
        loss = gluon.loss._apply_weighting(F, loss, self._weight, sample_weight)
        return F.mean(loss, axis=self._batch_axis, exclude=True)

Sebastian Nickels's avatar
Sebastian Nickels committed
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
class SoftmaxCrossEntropyLossIgnoreIndices(gluon.loss.Loss):
    def __init__(self, axis=-1, ignore_indices=[], sparse_label=True, from_logits=False, weight=None, batch_axis=0, **kwargs):
        super(SoftmaxCrossEntropyLossIgnoreIndices, self).__init__(weight, batch_axis, **kwargs)
        self._axis = axis
        self._ignore_indices = ignore_indices
        self._sparse_label = sparse_label
        self._from_logits = from_logits

    def hybrid_forward(self, F, pred, label, sample_weight=None):
        log_softmax = F.log_softmax
        pick = F.pick
        if not self._from_logits:
            pred = log_softmax(pred, self._axis)
        if self._sparse_label:
            loss = -pick(pred, label, axis=self._axis, keepdims=True)
        else:
            label = _reshape_like(F, label, pred)
            loss = -(pred * label).sum(axis=self._axis, keepdims=True)
        #loss = _apply_weighting(F, loss, self._weight, sample_weight)
        # ignore some indices for loss, e.g. <pad> tokens in NLP applications
        for i in self._ignore_indices:
            loss = loss * mx.nd.logical_not(mx.nd.equal(mx.nd.argmax(pred, axis=1), mx.nd.ones_like(mx.nd.argmax(pred, axis=1))*i))
        return loss.mean(axis=self._batch_axis, exclude=True)

Sebastian Nickels's avatar
Sebastian Nickels committed
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
@mx.metric.register
class BLEU(mx.metric.EvalMetric):
    N = 4

    def __init__(self, exclude=None, name='bleu', output_names=None, label_names=None):
        super(BLEU, self).__init__(name=name, output_names=output_names, label_names=label_names)

        self._exclude = exclude or []

        self._match_counts = [0 for _ in range(self.N)]
        self._counts = [0 for _ in range(self.N)]

        self._size_ref = 0
        self._size_hyp = 0

    def update(self, labels, preds):
        labels, preds = mx.metric.check_label_shapes(labels, preds, True)

        new_labels = self._convert(labels)
        new_preds = self._convert(preds)

        for label, pred in zip(new_labels, new_preds):
            reference = [word for word in label if word not in self._exclude]
            hypothesis = [word for word in pred if word not in self._exclude]

            self._size_ref += len(reference)
            self._size_hyp += len(hypothesis)

            for n in range(self.N):
                reference_ngrams = self._get_ngrams(reference, n + 1)
                hypothesis_ngrams = self._get_ngrams(hypothesis, n + 1)

                match_count = 0

                for ngram in hypothesis_ngrams:
                    if ngram in reference_ngrams:
                        reference_ngrams.remove(ngram)

                        match_count += 1

                self._match_counts[n] += match_count
                self._counts[n] += len(hypothesis_ngrams)

    def get(self):
        precisions = [sys.float_info.min for n in range(self.N)]

        i = 1

        for n in range(self.N):
            match_counts = self._match_counts[n]
            counts = self._counts[n]

            if counts != 0:
                if match_counts == 0:
                    i *= 2
                    match_counts = 1 / i

118
119
                if (match_counts / counts) > 0:
                    precisions[n] = match_counts / counts
Sebastian Nickels's avatar
Sebastian Nickels committed
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

        bleu = self._get_brevity_penalty() * math.exp(sum(map(math.log, precisions)) / self.N)

        return (self.name, bleu)

    def calculate(self):
        precisions = [sys.float_info.min for n in range(self.N)]

        i = 1

        for n in range(self.N):
            match_counts = self._match_counts[n]
            counts = self._counts[n]

            if counts != 0:
                if match_counts == 0:
                    i *= 2
                    match_counts = 1 / i

                precisions[n] = match_counts / counts

        return self._get_brevity_penalty() * math.exp(sum(map(math.log, precisions)) / self.N)

    def _get_brevity_penalty(self):
        if self._size_hyp >= self._size_ref:
            return 1
        else:
            return math.exp(1 - (self._size_ref / self._size_hyp))

    @staticmethod
    def _get_ngrams(sentence, n):
        ngrams = []

        if len(sentence) >= n:
            for i in range(len(sentence) - n + 1):
                ngrams.append(sentence[i:i+n])

        return ngrams

    @staticmethod
    def _convert(nd_list):
        if len(nd_list) == 0:
            return []

        new_list = [[] for _ in range(nd_list[0].shape[0])]

        for element in nd_list:
            for i in range(element.shape[0]):
                new_list[i].append(element[i].asscalar())

        return new_list
Christian Fuß's avatar
Christian Fuß committed
171

Sebastian Nickels's avatar
Sebastian Nickels committed
172
173


174
class CNNSupervisedTrainer_VGG16:
175
    def __init__(self, data_loader, net_constructor):
Nicola Gatto's avatar
Nicola Gatto committed
176
177
        self._data_loader = data_loader
        self._net_creator = net_constructor
178
        self._networks = {}
Nicola Gatto's avatar
Nicola Gatto committed
179
180
181
182

    def train(self, batch_size=64,
              num_epoch=10,
              eval_metric='acc',
Sebastian Nickels's avatar
Sebastian Nickels committed
183
              eval_metric_params={},
Eyüp Harputlu's avatar
Eyüp Harputlu committed
184
185
              loss ='softmax_cross_entropy',
              loss_params={},
Nicola Gatto's avatar
Nicola Gatto committed
186
187
188
189
190
              optimizer='adam',
              optimizer_params=(('learning_rate', 0.001),),
              load_checkpoint=True,
              context='gpu',
              checkpoint_period=5,
191
              save_attention_image=False,
Nicola Gatto's avatar
Nicola Gatto committed
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
              normalize=True):
        if context == 'gpu':
            mx_context = mx.gpu()
        elif context == 'cpu':
            mx_context = mx.cpu()
        else:
            logging.error("Context argument is '" + context + "'. Only 'cpu' and 'gpu are valid arguments'.")

        if 'weight_decay' in optimizer_params:
            optimizer_params['wd'] = optimizer_params['weight_decay']
            del optimizer_params['weight_decay']
        if 'learning_rate_decay' in optimizer_params:
            min_learning_rate = 1e-08
            if 'learning_rate_minimum' in optimizer_params:
                min_learning_rate = optimizer_params['learning_rate_minimum']
                del optimizer_params['learning_rate_minimum']
            optimizer_params['lr_scheduler'] = mx.lr_scheduler.FactorScheduler(
                                                   optimizer_params['step_size'],
                                                   factor=optimizer_params['learning_rate_decay'],
                                                   stop_factor_lr=min_learning_rate)
            del optimizer_params['step_size']
            del optimizer_params['learning_rate_decay']

215
216
        train_batch_size = batch_size
        test_batch_size = batch_size
Nicola Gatto's avatar
Nicola Gatto committed
217

Sebastian Nickels's avatar
Merge    
Sebastian Nickels committed
218
        train_iter, train_test_iter, test_iter, data_mean, data_std, train_images, test_images = self._data_loader.load_data(train_batch_size, test_batch_size)
219
220
221
222
223

        if normalize:
            self._net_creator.construct(context=mx_context, data_mean=data_mean, data_std=data_std)
        else:
            self._net_creator.construct(context=mx_context)
Nicola Gatto's avatar
Nicola Gatto committed
224
225
226
227
228
229
230
231

        begin_epoch = 0
        if load_checkpoint:
            begin_epoch = self._net_creator.load(mx_context)
        else:
            if os.path.isdir(self._net_creator._model_dir_):
                shutil.rmtree(self._net_creator._model_dir_)

232
        self._networks = self._net_creator.networks
Nicola Gatto's avatar
Nicola Gatto committed
233
234
235
236
237
238
239

        try:
            os.makedirs(self._net_creator._model_dir_)
        except OSError:
            if not os.path.isdir(self._net_creator._model_dir_):
                raise

Sebastian Nickels's avatar
Sebastian Nickels committed
240
        trainers = [mx.gluon.Trainer(network.collect_params(), optimizer, optimizer_params) for network in self._networks.values() if len(network.collect_params().values()) != 0]
Nicola Gatto's avatar
Nicola Gatto committed
241

Eyüp Harputlu's avatar
Eyüp Harputlu committed
242
243
        margin = loss_params['margin'] if 'margin' in loss_params else 1.0
        sparseLabel = loss_params['sparse_label'] if 'sparse_label' in loss_params else True
244
245
246
        #if loss == 'softmax_cross_entropy':
        #    fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
        #    loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(from_logits=fromLogits, sparse_label=sparseLabel)
Eyüp Harputlu's avatar
Eyüp Harputlu committed
247
248
        if loss == 'softmax_cross_entropy':
            fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
249
250
            ignore_indices = [2]
            loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel)
Eyüp Harputlu's avatar
Eyüp Harputlu committed
251
        elif loss == 'sigmoid_binary_cross_entropy':
Nicola Gatto's avatar
Nicola Gatto committed
252
            loss_function = mx.gluon.loss.SigmoidBinaryCrossEntropyLoss()
Eyüp Harputlu's avatar
Eyüp Harputlu committed
253
254
255
        elif loss == 'cross_entropy':
            loss_function = CrossEntropyLoss(sparse_label=sparseLabel)
        elif loss == 'l2':
Nicola Gatto's avatar
Nicola Gatto committed
256
            loss_function = mx.gluon.loss.L2Loss()
Eyüp Harputlu's avatar
Eyüp Harputlu committed
257
        elif loss == 'l1':
Nicola Gatto's avatar
Nicola Gatto committed
258
            loss_function = mx.gluon.loss.L2Loss()
Eyüp Harputlu's avatar
Eyüp Harputlu committed
259
260
261
262
263
264
265
266
267
268
269
270
271
        elif loss == 'huber':
            rho = loss_params['rho'] if 'rho' in loss_params else 1
            loss_function = mx.gluon.loss.HuberLoss(rho=rho)
        elif loss == 'hinge':
            loss_function = mx.gluon.loss.HingeLoss(margin=margin)
        elif loss == 'squared_hinge':
            loss_function = mx.gluon.loss.SquaredHingeLoss(margin=margin)
        elif loss == 'logistic':
            labelFormat = loss_params['label_format'] if 'label_format' in loss_params else 'signed'
            loss_function = mx.gluon.loss.LogisticLoss(label_format=labelFormat)
        elif loss == 'kullback_leibler':
            fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else True
            loss_function = mx.gluon.loss.KLDivLoss(from_logits=fromLogits)
Eyüp Harputlu's avatar
Eyüp Harputlu committed
272
273
        elif loss == 'log_cosh':
            loss_function = LogCoshLoss()
Eyüp Harputlu's avatar
Eyüp Harputlu committed
274
275
        else:
            logging.error("Invalid loss parameter.")
Nicola Gatto's avatar
Nicola Gatto committed
276
277
278
279
280
281
282

        speed_period = 50
        tic = None

        for epoch in range(begin_epoch, begin_epoch + num_epoch):
            train_iter.reset()
            for batch_i, batch in enumerate(train_iter):
283
284
                with autograd.record():
                    labels = [batch.label[i].as_in_context(mx_context) for i in range(1)]
285

286
287
288
                    data_ = batch.data[0].as_in_context(mx_context)

                    predictions_ = mx.nd.zeros((train_batch_size, 1000,), ctx=mx_context)
Christian Fuß's avatar
Christian Fuß committed
289

290

291
                    lossList = []
292

293
                    predictions_ = self._networks[0](data_)
294
295

                    lossList.append(loss_function(predictions_, labels[0]))
296

297
298
299
                    loss = 0
                    for element in lossList:
                        loss = loss + element
Nicola Gatto's avatar
Nicola Gatto committed
300
301

                loss.backward()
302
303
304

                for trainer in trainers:
                    trainer.step(batch_size)
Nicola Gatto's avatar
Nicola Gatto committed
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320

                if tic is None:
                    tic = time.time()
                else:
                    if batch_i % speed_period == 0:
                        try:
                            speed = speed_period * batch_size / (time.time() - tic)
                        except ZeroDivisionError:
                            speed = float("inf")

                        logging.info("Epoch[%d] Batch[%d] Speed: %.2f samples/sec" % (epoch, batch_i, speed))

                        tic = time.time()

            tic = None

321
            train_test_iter.reset()
Sebastian Nickels's avatar
Sebastian Nickels committed
322
            metric = mx.metric.create(eval_metric, **eval_metric_params)
323
            for batch_i, batch in enumerate(train_test_iter):
324
                if True: 
325
                    labels = [batch.label[i].as_in_context(mx_context) for i in range(1)]
326

327
                    data_ = batch.data[0].as_in_context(mx_context)
328

329
                    predictions_ = mx.nd.zeros((test_batch_size, 1000,), ctx=mx_context)
330

331

332
                    outputs = []
333
                    attentionList=[]
334
                    predictions_ = self._networks[0](data_)
335

336
                    outputs.append(predictions_)
337

338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357

                    if save_attention_image == "True":
                        import matplotlib.pyplot as plt
                        logging.getLogger('matplotlib').setLevel(logging.ERROR)

                        plt.clf()
                        fig = plt.figure(figsize=(10,10))
                        max_length = len(labels)-1

                        if(os.path.isfile('src/test/resources/training_data/Show_attend_tell/dict.pkl')):
                            with open('src/test/resources/training_data/Show_attend_tell/dict.pkl', 'rb') as f:
                                dict = pickle.load(f)

                        for l in range(max_length):
                            attention = attentionList[l]
                            attention = mx.nd.slice_axis(attention, axis=0, begin=0, end=1)
                            attention = mx.nd.squeeze(attention)
                            attention_resized = np.resize(attention.asnumpy(), (8, 8))
                            ax = fig.add_subplot(max_length//3, max_length//4, l+1)
                            ax.set_title(dict[int(labels[l+1][0].asscalar())])
358
                            img = ax.imshow(train_images[0+test_batch_size*(batch_i)].transpose(1,2,0))
359
360
361
362
363
364
365
366
367
368
                            ax.imshow(attention_resized, cmap='gray', alpha=0.6, extent=img.get_extent())


                        plt.tight_layout()
                        target_dir = 'target/attention_images'
                        if not os.path.exists(target_dir):
                                    os.makedirs(target_dir)
                        plt.savefig(target_dir + '/attention_train.png')
                        plt.close()

369
                predictions = []
370
                for output_name in outputs:
Sebastian Nickels's avatar
Sebastian Nickels committed
371
                    if mx.nd.shape_array(mx.nd.squeeze(output_name)).size > 1:
372
373
374
                        predictions.append(mx.nd.argmax(output_name, axis=1))
                    else:
                        predictions.append(output_name)
375
376

                metric.update(preds=predictions, labels=labels)
Nicola Gatto's avatar
Nicola Gatto committed
377
378
379
            train_metric_score = metric.get()[1]

            test_iter.reset()
Sebastian Nickels's avatar
Sebastian Nickels committed
380
            metric = mx.metric.create(eval_metric, **eval_metric_params)
Nicola Gatto's avatar
Nicola Gatto committed
381
            for batch_i, batch in enumerate(test_iter):
382
                if True: 
383
                    labels = [batch.label[i].as_in_context(mx_context) for i in range(1)]
384

385
                    data_ = batch.data[0].as_in_context(mx_context)
386

387
                    predictions_ = mx.nd.zeros((test_batch_size, 1000,), ctx=mx_context)
388

389

390
                    outputs = []
391
                    attentionList=[]
392
                    predictions_ = self._networks[0](data_)
393

394
                    outputs.append(predictions_)
395

396
397
398
399
400
401
402
403
404
405
406
407
408
409

                    if save_attention_image == "True":
                        plt.clf()
                        fig = plt.figure(figsize=(10,10))
                        max_length = len(labels)-1

                        for l in range(max_length):
                            attention = attentionList[l]
                            attention = mx.nd.slice_axis(attention, axis=2, begin=0, end=1)
                            attention = mx.nd.slice_axis(attention, axis=0, begin=0, end=1)
                            attention = mx.nd.squeeze(attention)
                            attention_resized = np.resize(attention.asnumpy(), (8, 8))
                            ax = fig.add_subplot(max_length//3, max_length//4, l+1)
                            ax.set_title(dict[int(mx.nd.slice_axis(mx.nd.argmax(outputs[l+1], axis=1), axis=0, begin=0, end=1).asscalar())])
410
                            img = ax.imshow(test_images[0+test_batch_size*(batch_i)].transpose(1,2,0))
411
412
413
414
415
416
417
                            ax.imshow(attention_resized, cmap='gray', alpha=0.6, extent=img.get_extent())


                        plt.tight_layout()
                        plt.savefig(target_dir + '/attention_test.png')
                        plt.close()

418
                predictions = []
419
                for output_name in outputs:
Sebastian Nickels's avatar
Sebastian Nickels committed
420
                    if mx.nd.shape_array(mx.nd.squeeze(output_name)).size > 1:
421
422
423
424
                        predictions.append(mx.nd.argmax(output_name, axis=1))
                    #ArgMax already applied
                    else:
                        predictions.append(output_name)
425

426
                metric.update(preds=predictions, labels=labels)
Nicola Gatto's avatar
Nicola Gatto committed
427
428
429
430
            test_metric_score = metric.get()[1]

            logging.info("Epoch[%d] Train: %f, Test: %f" % (epoch, train_metric_score, test_metric_score))

431

Nicola Gatto's avatar
Nicola Gatto committed
432
            if (epoch - begin_epoch) % checkpoint_period == 0:
433
434
                for i, network in self._networks.items():
                    network.save_parameters(self.parameter_path(i) + '-' + str(epoch).zfill(4) + '.params')
Nicola Gatto's avatar
Nicola Gatto committed
435

436
437
438
        for i, network in self._networks.items():
            network.save_parameters(self.parameter_path(i) + '-' + str(num_epoch + begin_epoch).zfill(4) + '.params')
            network.export(self.parameter_path(i) + '_newest', epoch=0)
Nicola Gatto's avatar
Nicola Gatto committed
439

440
    def parameter_path(self, index):
Bernhard Rumpe's avatar
BR-sy    
Bernhard Rumpe committed
441
        return self._net_creator._model_dir_ + self._net_creator._model_prefix_ + '_' + str(index)