Skip to content

Necks#

Necks reshape the output of an encoder into a format suitable for the decoder. By combining different necks, you can combine any backbone with any decoder.

terratorch.models.necks.Neck #

Bases: ABC, Module

Base class for Neck

A neck must must implement self.process_channel_list which returns the new channel list.

Source code in terratorch/models/necks.py 
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
class Neck(ABC, nn.Module):
    """Base class for Neck

    A neck must must implement `self.process_channel_list` which returns the new channel list.
    """

    def __init__(self, channel_list: list[int]) -> None:
        super().__init__()
        self.channel_list = channel_list

    @abstractmethod
    def process_channel_list(self, channel_list: list[int]) -> list[int]:
        return channel_list

    @abstractmethod
    def forward(self, channel_list: list[torch.Tensor]) -> list[torch.Tensor]: ...

terratorch.models.necks.SelectIndices #

Bases: Neck

Source code in terratorch/models/necks.py 
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
@TERRATORCH_NECK_REGISTRY.register
class SelectIndices(Neck):
    def __init__(self, channel_list: list[int], indices: list[int]):
        """Select indices from the embedding list

        Args:
            indices (list[int]): list of indices to select.
        """
        super().__init__(channel_list)
        self.indices = indices

    def forward(self, features: list[torch.Tensor]) -> list[torch.Tensor]:
        features = [features[i] for i in self.indices]
        return features

    def process_channel_list(self, channel_list: list[int]) -> list[int]:
        channel_list = [channel_list[i] for i in self.indices]
        return channel_list

__init__(channel_list, indices) #

Select indices from the embedding list

Parameters:

Name Type Description Default
indices list[int]

list of indices to select.

 required
Source code in terratorch/models/necks.py 
36
37
38
39
40
41
42
43
def __init__(self, channel_list: list[int], indices: list[int]):
    """Select indices from the embedding list

    Args:
        indices (list[int]): list of indices to select.
    """
    super().__init__(channel_list)
    self.indices = indices

terratorch.models.necks.ReshapeTokensToImage #

Bases: Neck

Source code in terratorch/models/necks.py 
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
@TERRATORCH_NECK_REGISTRY.register
class ReshapeTokensToImage(Neck):
    def __init__(
        self, channel_list: list[int], remove_cls_token=True, effective_time_dim: int = 1, h: int | None = None
    ):
        """Reshape output of transformer encoder so it can be passed to a conv net.

        Args:
            remove_cls_token (bool, optional): Whether to remove the cls token from the first position.
                Defaults to True.
            effective_time_dim (int, optional): The effective temporal dimension the transformer processes.
                For a ViT, his will be given by `num_frames // tubelet size`. This is used to determine
                the temporal dimension of the embedding, which is concatenated with the embedding dimension.
                For example:
                - A model which processes 1 frame with a tubelet size of 1 has an effective_time_dim of 1.
                    The embedding produced by this model has embedding size embed_dim * 1.
                - A model which processes 3 frames with a tubelet size of 1 has an effective_time_dim of 3.
                    The embedding produced by this model has embedding size embed_dim * 3.
                - A model which processes 12 frames with a tubelet size of 4 has an effective_time_dim of 3.
                    The embedding produced by this model has an embedding size embed_dim * 3.
                Defaults to 1.
            h (int | None):
                You can choose a value for the height of the reshaped image.
                The embedding size will be implicitly discovered from it.
        """
        super().__init__(channel_list)
        self.remove_cls_token = remove_cls_token
        self.effective_time_dim = effective_time_dim
        self.h = h

    def collapse_dims(self, x):
        """
        When the encoder output has more than 3 dimensions, is necessary to
        reshape it.
        """
        shape = x.shape
        batch = x.shape[0]
        e = x.shape[-1]
        collapsed_dim = np.prod(x.shape[1:-1])

        return x.reshape(batch, collapsed_dim, e)

    @staticmethod
    def is_prime(n):
        if n <= 1:
            return False
        for i in range(2, int(math.sqrt(n)) + 1):
            if n % i == 0:
                return False
        return True

    def factorize_to_get_h(self, tokens_per_timestep):
        primes = [2, 3, 5, 7, 11]
        j = primes[0]
        i = 0
        value = tokens_per_timestep
        dividers = []
        status = 0

        while not status:
            if self.is_prime(value):
                status = 1
            else:
                if value % j == 0:
                    value //= j
                    dividers.append(j)
                else:
                    i += 1
                    j = primes[i]

                status = 0

        return int(np.prod(dividers) / 2)

    def forward(self, features: list[torch.Tensor]) -> list[torch.Tensor]:
        out = []
        for x in features:
            if self.remove_cls_token:
                x_no_token = x[:, 1:, :]
            else:
                x_no_token = x
            x_no_token = self.collapse_dims(x_no_token)
            number_of_tokens = x_no_token.shape[1]
            tokens_per_timestep = number_of_tokens // self.effective_time_dim

            # Adaptation to use non-square images
            h = self.h or math.sqrt(tokens_per_timestep)
            if h - int(h) == 0:
                h = int(h)
            else:
                h = self.factorize_to_get_h(tokens_per_timestep)

            encoded = rearrange(
                x_no_token,
                "batch (t h w) e -> batch (t e) h w",
                batch=x_no_token.shape[0],
                t=self.effective_time_dim,
                h=h,
            )

            out.append(encoded)
        return out

    def process_channel_list(self, channel_list: list[int]) -> list[int]:
        return super().process_channel_list(channel_list)

__init__(channel_list, remove_cls_token=True, effective_time_dim=1, h=None) #

Reshape output of transformer encoder so it can be passed to a conv net.

Parameters:

Name Type Description Default
remove_cls_token bool

Whether to remove the cls token from the first position. Defaults to True.

 True
effective_time_dim int

The effective temporal dimension the transformer processes. For a ViT, his will be given by num_frames // tubelet size. This is used to determine the temporal dimension of the embedding, which is concatenated with the embedding dimension. For example: - A model which processes 1 frame with a tubelet size of 1 has an effective_time_dim of 1. The embedding produced by this model has embedding size embed_dim * 1. - A model which processes 3 frames with a tubelet size of 1 has an effective_time_dim of 3. The embedding produced by this model has embedding size embed_dim * 3. - A model which processes 12 frames with a tubelet size of 4 has an effective_time_dim of 3. The embedding produced by this model has an embedding size embed_dim * 3. Defaults to 1.

 1
h int | None

You can choose a value for the height of the reshaped image. The embedding size will be implicitly discovered from it.

 None
Source code in terratorch/models/necks.py 
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
def __init__(
    self, channel_list: list[int], remove_cls_token=True, effective_time_dim: int = 1, h: int | None = None
):
    """Reshape output of transformer encoder so it can be passed to a conv net.

    Args:
        remove_cls_token (bool, optional): Whether to remove the cls token from the first position.
            Defaults to True.
        effective_time_dim (int, optional): The effective temporal dimension the transformer processes.
            For a ViT, his will be given by `num_frames // tubelet size`. This is used to determine
            the temporal dimension of the embedding, which is concatenated with the embedding dimension.
            For example:
            - A model which processes 1 frame with a tubelet size of 1 has an effective_time_dim of 1.
                The embedding produced by this model has embedding size embed_dim * 1.
            - A model which processes 3 frames with a tubelet size of 1 has an effective_time_dim of 3.
                The embedding produced by this model has embedding size embed_dim * 3.
            - A model which processes 12 frames with a tubelet size of 4 has an effective_time_dim of 3.
                The embedding produced by this model has an embedding size embed_dim * 3.
            Defaults to 1.
        h (int | None):
            You can choose a value for the height of the reshaped image.
            The embedding size will be implicitly discovered from it.
    """
    super().__init__(channel_list)
    self.remove_cls_token = remove_cls_token
    self.effective_time_dim = effective_time_dim
    self.h = h

collapse_dims(x) #

When the encoder output has more than 3 dimensions, is necessary to reshape it.

Source code in terratorch/models/necks.py 
156
157
158
159
160
161
162
163
164
165
166
def collapse_dims(self, x):
    """
    When the encoder output has more than 3 dimensions, is necessary to
    reshape it.
    """
    shape = x.shape
    batch = x.shape[0]
    e = x.shape[-1]
    collapsed_dim = np.prod(x.shape[1:-1])

    return x.reshape(batch, collapsed_dim, e)

terratorch.models.necks.InterpolateToPyramidal #

Bases: Neck

Source code in terratorch/models/necks.py 
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
@TERRATORCH_NECK_REGISTRY.register
class InterpolateToPyramidal(Neck):
    def __init__(self, channel_list: list[int], scale_factor: int = 2, mode: str = "nearest"):
        """Spatially interpolate embeddings so that embedding[i - 1] is scale_factor times larger than embedding[i]

        Useful to make non-pyramidal backbones compatible with hierarachical ones
        Args:
            scale_factor (int): Amount to scale embeddings by each layer. Defaults to 2.
            mode (str): Interpolation mode to be passed to torch.nn.functional.interpolate. Defaults to 'nearest'.
        """
        super().__init__(channel_list)
        self.scale_factor = scale_factor
        self.mode = mode

    def forward(self, features: list[torch.Tensor]) -> list[torch.Tensor]:
        out = []
        scale_exponents = list(range(len(features), 0, -1))
        for x, exponent in zip(features, scale_exponents, strict=True):
            out.append(F.interpolate(x, scale_factor=self.scale_factor**exponent, mode=self.mode))

        return out

    def process_channel_list(self, channel_list: list[int]) -> list[int]:
        return super().process_channel_list(channel_list)

__init__(channel_list, scale_factor=2, mode='nearest') #

Spatially interpolate embeddings so that embedding[i - 1] is scale_factor times larger than embedding[i]

Useful to make non-pyramidal backbones compatible with hierarachical ones Args: scale_factor (int): Amount to scale embeddings by each layer. Defaults to 2. mode (str): Interpolation mode to be passed to torch.nn.functional.interpolate. Defaults to 'nearest'.

Source code in terratorch/models/necks.py 
75
76
77
78
79
80
81
82
83
84
85
def __init__(self, channel_list: list[int], scale_factor: int = 2, mode: str = "nearest"):
    """Spatially interpolate embeddings so that embedding[i - 1] is scale_factor times larger than embedding[i]

    Useful to make non-pyramidal backbones compatible with hierarachical ones
    Args:
        scale_factor (int): Amount to scale embeddings by each layer. Defaults to 2.
        mode (str): Interpolation mode to be passed to torch.nn.functional.interpolate. Defaults to 'nearest'.
    """
    super().__init__(channel_list)
    self.scale_factor = scale_factor
    self.mode = mode

terratorch.models.necks.LearnedInterpolateToPyramidal #

Bases: Neck

Use learned convolutions to transform the output of a non-pyramidal encoder into pyramidal ones

Always requires exactly 4 embeddings

Source code in terratorch/models/necks.py 
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
@TERRATORCH_NECK_REGISTRY.register
class LearnedInterpolateToPyramidal(Neck):
    """Use learned convolutions to transform the output of a non-pyramidal encoder into pyramidal ones

    Always requires exactly 4 embeddings
    """

    def __init__(self, channel_list: list[int]):
        super().__init__(channel_list)
        if len(channel_list) != 4:
            msg = "This class can only handle exactly 4 input embeddings"
            raise Exception(msg)
        self.fpn1 = nn.Sequential(
            nn.ConvTranspose2d(channel_list[0], channel_list[0] // 2, 2, 2),
            nn.BatchNorm2d(channel_list[0] // 2),
            nn.GELU(),
            nn.ConvTranspose2d(channel_list[0] // 2, channel_list[0] // 4, 2, 2),
        )
        self.fpn2 = nn.Sequential(nn.ConvTranspose2d(channel_list[1], channel_list[1] // 2, 2, 2))
        self.fpn3 = nn.Sequential(nn.Identity())
        self.fpn4 = nn.Sequential(nn.MaxPool2d(kernel_size=2, stride=2))
        self.embedding_dim = [channel_list[0] // 4, channel_list[1] // 2, channel_list[2], channel_list[3]]

    def forward(self, features: list[torch.Tensor]) -> list[torch.Tensor]:
        scaled_inputs = []
        scaled_inputs.append(self.fpn1(features[0]))
        scaled_inputs.append(self.fpn2(features[1]))
        scaled_inputs.append(self.fpn3(features[2]))
        scaled_inputs.append(self.fpn4(features[3]))
        return scaled_inputs

    def process_channel_list(self, channel_list: list[int] = None) -> list[int]:
        return [channel_list[0] // 4, channel_list[1] // 2, channel_list[2], channel_list[3]]

terratorch.models.necks.PermuteDims #

Bases: Neck

Source code in terratorch/models/necks.py 
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
@TERRATORCH_NECK_REGISTRY.register
class PermuteDims(Neck):
    def __init__(self, channel_list: list[int], new_order: list[int]):
        """Permute dimensions of each element in the embedding list

        Args:
            new_order (list[int]): list of indices to be passed to tensor.permute()
        """
        super().__init__(channel_list)
        self.new_order = new_order

    def forward(self, features: list[torch.Tensor]) -> list[torch.Tensor]:
        features = [feat.permute(*self.new_order).contiguous() for feat in features]
        return features

    def process_channel_list(self, channel_list: list[int]) -> list[int]:
        return super().process_channel_list(channel_list)

__init__(channel_list, new_order) #

Permute dimensions of each element in the embedding list

Parameters:

Name Type Description Default
new_order list[int]

list of indices to be passed to tensor.permute()

 required
Source code in terratorch/models/necks.py 
56
57
58
59
60
61
62
63
def __init__(self, channel_list: list[int], new_order: list[int]):
    """Permute dimensions of each element in the embedding list

    Args:
        new_order (list[int]): list of indices to be passed to tensor.permute()
    """
    super().__init__(channel_list)
    self.new_order = new_order

terratorch.models.necks.MaxpoolToPyramidal #

Bases: Neck

Source code in terratorch/models/necks.py 
 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
@TERRATORCH_NECK_REGISTRY.register
class MaxpoolToPyramidal(Neck):
    def __init__(self, channel_list: list[int], kernel_size: int = 2):
        """Spatially downsample embeddings so that embedding[i - 1] is scale_factor times smaller than embedding[i]

        Useful to make non-pyramidal backbones compatible with hierarachical ones
        Args:
            kernel_size (int). Base kernel size to use for maxpool. Defaults to 2.
        """
        super().__init__(channel_list)
        self.kernel_size = kernel_size

    def forward(self, features: list[torch.Tensor]) -> list[torch.Tensor]:
        out = []
        scale_exponents = list(range(len(features)))
        for x, exponent in zip(features, scale_exponents, strict=True):
            if exponent == 0:
                out.append(x.clone())
            else:
                out.append(F.max_pool2d(x, kernel_size=self.kernel_size**exponent))

        return out

    def process_channel_list(self, channel_list: list[int]) -> list[int]:
        return super().process_channel_list(channel_list)

__init__(channel_list, kernel_size=2) #

Spatially downsample embeddings so that embedding[i - 1] is scale_factor times smaller than embedding[i]

Useful to make non-pyramidal backbones compatible with hierarachical ones Args: kernel_size (int). Base kernel size to use for maxpool. Defaults to 2.

Source code in terratorch/models/necks.py 
101
102
103
104
105
106
107
108
109
def __init__(self, channel_list: list[int], kernel_size: int = 2):
    """Spatially downsample embeddings so that embedding[i - 1] is scale_factor times smaller than embedding[i]

    Useful to make non-pyramidal backbones compatible with hierarachical ones
    Args:
        kernel_size (int). Base kernel size to use for maxpool. Defaults to 2.
    """
    super().__init__(channel_list)
    self.kernel_size = kernel_size

terratorch.models.necks.AddBottleneckLayer #

Bases: Neck

Add a layer that reduces the channel dimension of the final embedding by half, and concatenates it

Useful for compatibility with some smp decoders.

Source code in terratorch/models/necks.py 
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
@TERRATORCH_NECK_REGISTRY.register
class AddBottleneckLayer(Neck):
    """Add a layer that reduces the channel dimension of the final embedding by half, and concatenates it

    Useful for compatibility with some smp decoders.
    """

    def __init__(self, channel_list: list[int]):
        super().__init__(channel_list)
        self.bottleneck = nn.Conv2d(channel_list[-1], channel_list[-1] // 2, kernel_size=1)

    def forward(self, features: list[torch.Tensor]) -> list[torch.Tensor]:
        new_embedding = self.bottleneck(features[-1])
        features.append(new_embedding)
        return features

    def process_channel_list(self, channel_list: list[int]) -> list[int]:
        return [*channel_list, channel_list[-1] // 2]