Cropping3D 層
Cropping3D 類別
keras.layers.Cropping3D(
cropping=((1, 1), (1, 1), (1, 1)), data_format=None, **kwargs
)
用於 3D 資料（例如空間或時空）的裁剪層。
範例
>>> input_shape = (2, 28, 28, 10, 3)
>>> x = np.arange(np.prod(input_shape)).reshape(input_shape)
>>> y = keras.layers.Cropping3D(cropping=(2, 4, 2))(x)
>>> y.shape
(2, 24, 20, 6, 3)
引數
- cropping: 整數，或 3 個整數的元組，或 3 個 2 個整數元組的元組。
- 如果是整數：則對深度、高度和寬度應用相同的對稱裁剪。
- 如果是 3 個整數的元組：則解釋為深度、高度和寬度的三個不同對稱裁剪值：
(symmetric_dim1_crop, symmetric_dim2_crop, symmetric_dim3_crop)。 - 如果是 3 個 2 個整數元組的元組：則解釋為
((left_dim1_crop, right_dim1_crop), (left_dim2_crop, right_dim2_crop), (left_dim3_crop, right_dim3_crop))。
- data_format: 字串，可以是
"channels_last"(預設) 或"channels_first"。輸入中維度的順序。"channels_last"對應於形狀為(batch_size, spatial_dim1, spatial_dim2, spatial_dim3, channels)的輸入，而"channels_first"對應於形狀為(batch_size, channels, spatial_dim1, spatial_dim2, spatial_dim3)的輸入。未指定時，使用在您的 Keras 設定檔~/.keras/keras.json(如果存在) 中找到的image_data_format值。預設為"channels_last"。
輸入形狀
5D 張量，形狀為： - 如果 data_format 為 "channels_last": (batch_size, first_axis_to_crop, second_axis_to_crop, third_axis_to_crop, channels) - 如果 data_format 為 "channels_first": (batch_size, channels, first_axis_to_crop, second_axis_to_crop, third_axis_to_crop)
輸出形狀
5D 張量，形狀為： - 如果 data_format 為 "channels_last": (batch_size, first_cropped_axis, second_cropped_axis, third_cropped_axis, channels) - 如果 data_format 為 "channels_first": (batch_size, channels, first_cropped_axis, second_cropped_axis, third_cropped_axis)