File size: 8,441 Bytes
3f50570
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
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
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
import math
from typing import Tuple
import torch
import torch.nn as nn
from torchaudio.transforms import SpecAugment
from torch import Tensor
from torchvision.transforms import functional as F


class AugmentLayer(nn.Module):
    def __init__(self, cfg):
        super().__init__()
        self.cfg = cfg

        # Initialize MixUp
        self.mixup = MixUp(
            alpha=cfg.augment.mixup_alpha,
            num_classes=cfg.num_classes,
            p=cfg.augment.mixup_p,
            inplace=True,
        )

        # Initialize other augmentations
        self.time_freq_mask = SpecAugment(
            n_time_masks=cfg.augment.n_time_masks,
            time_mask_param=cfg.augment.time_mask_param,
            n_freq_masks=cfg.augment.n_freq_masks,
            freq_mask_param=cfg.augment.freq_mask_param,
            p=cfg.augment.time_freq_mask_p,
            zero_masking=True,
        )

    def forward(self, spec, y=None):
        # Apply MixUp or CutMix with RandomChoice
        if y is not None:
            # img = spec.unsqueeze(1)  # shape: (batch_size, 1, n_mels, n_frames)
            spec, y = self.mixup(spec, y)
            # spec = img.squeeze(1)  # shape: (batch_size, n_mels, n_frames)

        # Apply TimeMasking and FrequencyMasking
        spec = self.time_freq_mask(spec)
        return spec, y


class MixUp(torch.nn.Module):
    """Randomly apply MixUp to the provided batch and targets.
    The class implements the data augmentations as described in the paper
    `"mixup: Beyond Empirical Risk Minimization" <https://arxiv.org/abs/1710.09412>`_.

    Args:
        num_classes (int): number of classes used for one-hot encoding.
        p (float): probability of the batch being transformed. Default value is 0.5.
        alpha (float): hyperparameter of the Beta distribution used for mixup.
            Default value is 1.0.
        inplace (bool): boolean to make this transform inplace. Default set to False.
    """

    def __init__(
        self,
        num_classes: int,
        p: float = 0.5,
        alpha: float = 1.0,
        inplace: bool = False,
    ) -> None:
        super().__init__()

        if num_classes < 1:
            raise ValueError(
                f"Please provide a valid positive value for the num_classes. Got num_classes={num_classes}"
            )

        if alpha <= 0:
            raise ValueError("Alpha param can't be zero.")

        self.num_classes = num_classes
        self.p = p
        self.alpha = alpha
        self.inplace = inplace

    def forward(self, batch: Tensor, target: Tensor) -> Tuple[Tensor, Tensor]:
        """
        Args:
            batch (Tensor): Float tensor of size (B, C, H, W)
            target (Tensor): Integer tensor of size (B, )

        Returns:
            Tensor: Randomly transformed batch.
        """
        if batch.ndim != 3 and batch.ndim != 2:
            raise ValueError(
                f"Batch ndim should be 3 (b, f, t) or 2 (b, n). Got {batch.ndim}"
            )
        if target.ndim != 1:
            raise ValueError(f"Target ndim should be 1. Got {target.ndim}")
        if not batch.is_floating_point():
            raise TypeError(f"Batch dtype should be a float tensor. Got {batch.dtype}.")
        if target.dtype != torch.int64 and self.num_classes > 1:
            raise TypeError(f"Target dtype should be torch.int64. Got {target.dtype}")

        if not self.inplace:
            batch = batch.clone()
            target = target.clone()

        if target.ndim == 1 and self.num_classes > 1:
            target = torch.nn.functional.one_hot(target, num_classes=self.num_classes)

        target = target.to(dtype=batch.dtype)

        if torch.rand(1).item() >= self.p:
            return batch, target

        # It's faster to roll the batch by one instead of shuffling it to create image pairs
        batch_rolled = batch.roll(1, 0)
        target_rolled = target.roll(1, 0)

        # Implemented as on mixup paper, page 3.
        lambda_param = float(
            torch._sample_dirichlet(torch.tensor([self.alpha, self.alpha]))[0]
        )
        batch_rolled.mul_(1.0 - lambda_param)
        batch.mul_(lambda_param).add_(batch_rolled)

        target_rolled.mul_(1.0 - lambda_param)
        target.mul_(lambda_param).add_(target_rolled)

        return batch, target

    def __repr__(self) -> str:
        s = (
            f"{self.__class__.__name__}("
            f"num_classes={self.num_classes}"
            f", p={self.p}"
            f", alpha={self.alpha}"
            f", inplace={self.inplace}"
            f")"
        )
        return s


# Todo: height of spec should be 1, adjust it for audio input (bs, n_samples)
class CutMix(torch.nn.Module):
    """Randomly apply CutMix to the provided batch and targets.
    The class implements the data augmentations as described in the paper
    `"CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features"
    <https://arxiv.org/abs/1905.04899>`_.

    Args:
        num_classes (int): number of classes used for one-hot encoding.
        p (float): probability of the batch being transformed. Default value is 0.5.
        alpha (float): hyperparameter of the Beta distribution used for cutmix.
            Default value is 1.0.
        inplace (bool): boolean to make this transform inplace. Default set to False.
    """

    def __init__(
        self,
        num_classes: int,
        p: float = 0.5,
        alpha: float = 1.0,
        inplace: bool = False,
    ) -> None:
        super().__init__()
        if num_classes < 1:
            raise ValueError(
                "Please provide a valid positive value for the num_classes."
            )
        if alpha <= 0:
            raise ValueError("Alpha param can't be zero.")

        self.num_classes = num_classes
        self.p = p
        self.alpha = alpha
        self.inplace = inplace

    def forward(self, batch: Tensor, target: Tensor) -> Tuple[Tensor, Tensor]:
        """
        Args:
            batch (Tensor): Float tensor of size (B, C, H, W)
            target (Tensor): Integer tensor of size (B, )

        Returns:
            Tensor: Randomly transformed batch.
        """
        if batch.ndim != 4:
            raise ValueError(f"Batch ndim should be 4. Got {batch.ndim}")
        if target.ndim != 1:
            raise ValueError(f"Target ndim should be 1. Got {target.ndim}")
        if not batch.is_floating_point():
            raise TypeError(f"Batch dtype should be a float tensor. Got {batch.dtype}.")
        if target.dtype != torch.int64 and self.num_classes > 1:
            raise TypeError(f"Target dtype should be torch.int64. Got {target.dtype}")

        if not self.inplace:
            batch = batch.clone()
            target = target.clone()

        if target.ndim == 1 and self.num_classes > 1:
            target = torch.nn.functional.one_hot(target, num_classes=self.num_classes)

        target = target.to(dtype=batch.dtype)

        if torch.rand(1).item() >= self.p:
            return batch, target

        # It's faster to roll the batch by one instead of shuffling it to create image pairs
        batch_rolled = batch.roll(1, 0)
        target_rolled = target.roll(1, 0)

        # Implemented as on cutmix paper, page 12 (with minor corrections on typos).
        lambda_param = float(
            torch._sample_dirichlet(torch.tensor([self.alpha, self.alpha]))[0]
        )
        _, H, W = F.get_dimensions(batch)

        r_x = torch.randint(W, (1,))
        r_y = torch.randint(H, (1,))

        r = 0.5 * math.sqrt(1.0 - lambda_param)
        r_w_half = int(r * W)
        r_h_half = int(r * H)

        x1 = int(torch.clamp(r_x - r_w_half, min=0))
        y1 = int(torch.clamp(r_y - r_h_half, min=0))
        x2 = int(torch.clamp(r_x + r_w_half, max=W))
        y2 = int(torch.clamp(r_y + r_h_half, max=H))

        batch[:, :, y1:y2, x1:x2] = batch_rolled[:, :, y1:y2, x1:x2]
        lambda_param = float(1.0 - (x2 - x1) * (y2 - y1) / (W * H))

        target_rolled.mul_(1.0 - lambda_param)
        target.mul_(lambda_param).add_(target_rolled)

        return batch, target

    def __repr__(self) -> str:
        s = (
            f"{self.__class__.__name__}("
            f"num_classes={self.num_classes}"
            f", p={self.p}"
            f", alpha={self.alpha}"
            f", inplace={self.inplace}"
            f")"
        )
        return s