Trait burn::tensor::ops::ModuleOps

pub trait ModuleOps<B>
where
    B: Backend,
{
    // Required methods
    fn conv2d(
        x: <B as Backend>::FloatTensorPrimitive<4>,
        weight: <B as Backend>::FloatTensorPrimitive<4>,
        bias: Option<<B as Backend>::FloatTensorPrimitive<1>>,
        options: ConvOptions<2>,
    ) -> <B as Backend>::FloatTensorPrimitive<4>;
    fn conv3d(
        x: <B as Backend>::FloatTensorPrimitive<5>,
        weight: <B as Backend>::FloatTensorPrimitive<5>,
        bias: Option<<B as Backend>::FloatTensorPrimitive<1>>,
        options: ConvOptions<3>,
    ) -> <B as Backend>::FloatTensorPrimitive<5>;
    fn conv_transpose2d(
        x: <B as Backend>::FloatTensorPrimitive<4>,
        weight: <B as Backend>::FloatTensorPrimitive<4>,
        bias: Option<<B as Backend>::FloatTensorPrimitive<1>>,
        options: ConvTransposeOptions<2>,
    ) -> <B as Backend>::FloatTensorPrimitive<4>;
    fn conv_transpose3d(
        x: <B as Backend>::FloatTensorPrimitive<5>,
        weight: <B as Backend>::FloatTensorPrimitive<5>,
        bias: Option<<B as Backend>::FloatTensorPrimitive<1>>,
        options: ConvTransposeOptions<3>,
    ) -> <B as Backend>::FloatTensorPrimitive<5>;
    fn avg_pool2d(
        x: <B as Backend>::FloatTensorPrimitive<4>,
        kernel_size: [usize; 2],
        stride: [usize; 2],
        padding: [usize; 2],
        count_include_pad: bool,
    ) -> <B as Backend>::FloatTensorPrimitive<4>;
    fn avg_pool2d_backward(
        x: <B as Backend>::FloatTensorPrimitive<4>,
        grad: <B as Backend>::FloatTensorPrimitive<4>,
        kernel_size: [usize; 2],
        stride: [usize; 2],
        padding: [usize; 2],
        count_include_pad: bool,
    ) -> <B as Backend>::FloatTensorPrimitive<4>;
    fn adaptive_avg_pool2d(
        x: <B as Backend>::FloatTensorPrimitive<4>,
        output_size: [usize; 2],
    ) -> <B as Backend>::FloatTensorPrimitive<4>;
    fn adaptive_avg_pool2d_backward(
        x: <B as Backend>::FloatTensorPrimitive<4>,
        grad: <B as Backend>::FloatTensorPrimitive<4>,
    ) -> <B as Backend>::FloatTensorPrimitive<4>;
    fn max_pool2d(
        x: <B as Backend>::FloatTensorPrimitive<4>,
        kernel_size: [usize; 2],
        stride: [usize; 2],
        padding: [usize; 2],
        dilation: [usize; 2],
    ) -> <B as Backend>::FloatTensorPrimitive<4>;
    fn max_pool2d_with_indices(
        x: <B as Backend>::FloatTensorPrimitive<4>,
        kernel_size: [usize; 2],
        stride: [usize; 2],
        padding: [usize; 2],
        dilation: [usize; 2],
    ) -> MaxPool2dWithIndices<B>;
    fn max_pool2d_with_indices_backward(
        x: <B as Backend>::FloatTensorPrimitive<4>,
        kernel_size: [usize; 2],
        stride: [usize; 2],
        padding: [usize; 2],
        dilation: [usize; 2],
        output_grad: <B as Backend>::FloatTensorPrimitive<4>,
        indices: <B as Backend>::IntTensorPrimitive<4>,
    ) -> MaxPool2dBackward<B>;
    fn interpolate(
        x: <B as Backend>::FloatTensorPrimitive<4>,
        output_size: [usize; 2],
        options: InterpolateOptions,
    ) -> <B as Backend>::FloatTensorPrimitive<4>;
    fn interpolate_backward(
        x: <B as Backend>::FloatTensorPrimitive<4>,
        grad: <B as Backend>::FloatTensorPrimitive<4>,
        output_size: [usize; 2],
        options: InterpolateOptions,
    ) -> <B as Backend>::FloatTensorPrimitive<4>;

    // Provided methods
    fn embedding(
        weights: <B as Backend>::FloatTensorPrimitive<2>,
        indices: <B as Backend>::IntTensorPrimitive<2>,
    ) -> <B as Backend>::FloatTensorPrimitive<3> { ... }
    fn embedding_backward(
        weights: <B as Backend>::FloatTensorPrimitive<2>,
        output_grad: <B as Backend>::FloatTensorPrimitive<3>,
        indices: <B as Backend>::IntTensorPrimitive<2>,
    ) -> <B as Backend>::FloatTensorPrimitive<2> { ... }
    fn conv1d(
        x: <B as Backend>::FloatTensorPrimitive<3>,
        weight: <B as Backend>::FloatTensorPrimitive<3>,
        bias: Option<<B as Backend>::FloatTensorPrimitive<1>>,
        options: ConvOptions<1>,
    ) -> <B as Backend>::FloatTensorPrimitive<3> { ... }
    fn conv1d_backward(
        x: <B as Backend>::FloatTensorPrimitive<3>,
        weight: <B as Backend>::FloatTensorPrimitive<3>,
        bias: Option<<B as Backend>::FloatTensorPrimitive<1>>,
        output_grad: <B as Backend>::FloatTensorPrimitive<3>,
        options: ConvOptions<1>,
    ) -> Conv1dBackward<B> { ... }
    fn conv2d_backward(
        x: <B as Backend>::FloatTensorPrimitive<4>,
        weight: <B as Backend>::FloatTensorPrimitive<4>,
        bias: Option<<B as Backend>::FloatTensorPrimitive<1>>,
        output_grad: <B as Backend>::FloatTensorPrimitive<4>,
        options: ConvOptions<2>,
    ) -> Conv2dBackward<B> { ... }
    fn conv3d_backward(
        x: <B as Backend>::FloatTensorPrimitive<5>,
        weight: <B as Backend>::FloatTensorPrimitive<5>,
        bias: Option<<B as Backend>::FloatTensorPrimitive<1>>,
        output_grad: <B as Backend>::FloatTensorPrimitive<5>,
        options: ConvOptions<3>,
    ) -> Conv3dBackward<B> { ... }
    fn conv_transpose1d(
        x: <B as Backend>::FloatTensorPrimitive<3>,
        weight: <B as Backend>::FloatTensorPrimitive<3>,
        bias: Option<<B as Backend>::FloatTensorPrimitive<1>>,
        options: ConvTransposeOptions<1>,
    ) -> <B as Backend>::FloatTensorPrimitive<3> { ... }
    fn conv_transpose1d_backward(
        x: <B as Backend>::FloatTensorPrimitive<3>,
        weight: <B as Backend>::FloatTensorPrimitive<3>,
        bias: Option<<B as Backend>::FloatTensorPrimitive<1>>,
        output_grad: <B as Backend>::FloatTensorPrimitive<3>,
        options: ConvTransposeOptions<1>,
    ) -> Conv1dBackward<B> { ... }
    fn conv_transpose2d_backward(
        x: <B as Backend>::FloatTensorPrimitive<4>,
        weight: <B as Backend>::FloatTensorPrimitive<4>,
        bias: Option<<B as Backend>::FloatTensorPrimitive<1>>,
        output_grad: <B as Backend>::FloatTensorPrimitive<4>,
        options: ConvTransposeOptions<2>,
    ) -> Conv2dBackward<B> { ... }
    fn conv_transpose3d_backward(
        x: <B as Backend>::FloatTensorPrimitive<5>,
        weight: <B as Backend>::FloatTensorPrimitive<5>,
        bias: Option<<B as Backend>::FloatTensorPrimitive<1>>,
        output_grad: <B as Backend>::FloatTensorPrimitive<5>,
        options: ConvTransposeOptions<3>,
    ) -> Conv3dBackward<B> { ... }
    fn unfold4d(
        x: <B as Backend>::FloatTensorPrimitive<4>,
        kernel_size: [usize; 2],
        options: UnfoldOptions,
    ) -> <B as Backend>::FloatTensorPrimitive<3> { ... }
    fn avg_pool1d(
        x: <B as Backend>::FloatTensorPrimitive<3>,
        kernel_size: usize,
        stride: usize,
        padding: usize,
        count_include_pad: bool,
    ) -> <B as Backend>::FloatTensorPrimitive<3> { ... }
    fn avg_pool1d_backward(
        x: <B as Backend>::FloatTensorPrimitive<3>,
        grad: <B as Backend>::FloatTensorPrimitive<3>,
        kernel_size: usize,
        stride: usize,
        padding: usize,
        count_include_pad: bool,
    ) -> <B as Backend>::FloatTensorPrimitive<3> { ... }
    fn adaptive_avg_pool1d(
        x: <B as Backend>::FloatTensorPrimitive<3>,
        output_size: usize,
    ) -> <B as Backend>::FloatTensorPrimitive<3> { ... }
    fn adaptive_avg_pool1d_backward(
        x: <B as Backend>::FloatTensorPrimitive<3>,
        grad: <B as Backend>::FloatTensorPrimitive<3>,
    ) -> <B as Backend>::FloatTensorPrimitive<3> { ... }
    fn max_pool1d(
        x: <B as Backend>::FloatTensorPrimitive<3>,
        kernel_size: usize,
        stride: usize,
        padding: usize,
        dilation: usize,
    ) -> <B as Backend>::FloatTensorPrimitive<3> { ... }
    fn max_pool1d_with_indices(
        x: <B as Backend>::FloatTensorPrimitive<3>,
        kernel_size: usize,
        stride: usize,
        padding: usize,
        dilation: usize,
    ) -> MaxPool1dWithIndices<B> { ... }
    fn max_pool1d_with_indices_backward(
        x: <B as Backend>::FloatTensorPrimitive<3>,
        kernel_size: usize,
        stride: usize,
        padding: usize,
        dilation: usize,
        output_grad: <B as Backend>::FloatTensorPrimitive<3>,
        indices: <B as Backend>::IntTensorPrimitive<3>,
    ) -> MaxPool1dBackward<B> { ... }
}

Module operations trait.

Required Methods

fn conv2d( x: <B as Backend>::FloatTensorPrimitive<4>, weight: <B as Backend>::FloatTensorPrimitive<4>, bias: Option<<B as Backend>::FloatTensorPrimitive<1>>, options: ConvOptions<2>, ) -> <B as Backend>::FloatTensorPrimitive<4>

Two dimensional convolution.

Shapes

x: [batch_size, channels_in, height, width], weight: [channels_out, channels_in, kernel_size_1, kernel_size_2], bias: [channels_out],

fn conv3d( x: <B as Backend>::FloatTensorPrimitive<5>, weight: <B as Backend>::FloatTensorPrimitive<5>, bias: Option<<B as Backend>::FloatTensorPrimitive<1>>, options: ConvOptions<3>, ) -> <B as Backend>::FloatTensorPrimitive<5>

Three dimensional convolution.

Shapes

x: [batch_size, channels_in, depth, height, width], weight: [channels_out, channels_in, kernel_size_1, kernel_size_2, kernel_size_3], bias: [channels_out],

fn conv_transpose2d( x: <B as Backend>::FloatTensorPrimitive<4>, weight: <B as Backend>::FloatTensorPrimitive<4>, bias: Option<<B as Backend>::FloatTensorPrimitive<1>>, options: ConvTransposeOptions<2>, ) -> <B as Backend>::FloatTensorPrimitive<4>

Two dimensional transposed convolution.

Shapes

x: [batch_size, channels_in, height, width], weight: [channels_in, channels_out, kernel_size_1, kernel_size_2], bias: [channels_out],

fn conv_transpose3d( x: <B as Backend>::FloatTensorPrimitive<5>, weight: <B as Backend>::FloatTensorPrimitive<5>, bias: Option<<B as Backend>::FloatTensorPrimitive<1>>, options: ConvTransposeOptions<3>, ) -> <B as Backend>::FloatTensorPrimitive<5>

Three dimensional transposed convolution.

Shapes

x: [batch_size, channels_in, height, width], weight: [channels_in, channels_out, kernel_size_1, kernel_size_2, kernel_size_3], bias: [channels_out],

fn avg_pool2d( x: <B as Backend>::FloatTensorPrimitive<4>, kernel_size: [usize; 2], stride: [usize; 2], padding: [usize; 2], count_include_pad: bool, ) -> <B as Backend>::FloatTensorPrimitive<4>

Two dimensional avg pooling.

Shapes

x: [batch_size, channels, height, width],

fn avg_pool2d_backward( x: <B as Backend>::FloatTensorPrimitive<4>, grad: <B as Backend>::FloatTensorPrimitive<4>, kernel_size: [usize; 2], stride: [usize; 2], padding: [usize; 2], count_include_pad: bool, ) -> <B as Backend>::FloatTensorPrimitive<4>

Backward pass for the avg pooling 2d operation.

fn adaptive_avg_pool2d( x: <B as Backend>::FloatTensorPrimitive<4>, output_size: [usize; 2], ) -> <B as Backend>::FloatTensorPrimitive<4>

Two dimensional adaptive avg pooling.

Shapes

x: [batch_size, channels, height, width],

fn adaptive_avg_pool2d_backward( x: <B as Backend>::FloatTensorPrimitive<4>, grad: <B as Backend>::FloatTensorPrimitive<4>, ) -> <B as Backend>::FloatTensorPrimitive<4>

Backward pass for the adaptive avg pooling 2d operation.

fn max_pool2d( x: <B as Backend>::FloatTensorPrimitive<4>, kernel_size: [usize; 2], stride: [usize; 2], padding: [usize; 2], dilation: [usize; 2], ) -> <B as Backend>::FloatTensorPrimitive<4>

Two dimensional max pooling.

Shapes

x: [batch_size, channels, height, width],

fn max_pool2d_with_indices( x: <B as Backend>::FloatTensorPrimitive<4>, kernel_size: [usize; 2], stride: [usize; 2], padding: [usize; 2], dilation: [usize; 2], ) -> MaxPool2dWithIndices<B>

Two dimensional max pooling with indices.

Shapes

x: [batch_size, channels, height, width],

fn max_pool2d_with_indices_backward( x: <B as Backend>::FloatTensorPrimitive<4>, kernel_size: [usize; 2], stride: [usize; 2], padding: [usize; 2], dilation: [usize; 2], output_grad: <B as Backend>::FloatTensorPrimitive<4>, indices: <B as Backend>::IntTensorPrimitive<4>, ) -> MaxPool2dBackward<B>

Backward pass for the max pooling 2d operation.

fn interpolate( x: <B as Backend>::FloatTensorPrimitive<4>, output_size: [usize; 2], options: InterpolateOptions, ) -> <B as Backend>::FloatTensorPrimitive<4>

Down/up samples the input.

Shapes

x: [batch_size, channels, height, width],

fn interpolate_backward( x: <B as Backend>::FloatTensorPrimitive<4>, grad: <B as Backend>::FloatTensorPrimitive<4>, output_size: [usize; 2], options: InterpolateOptions, ) -> <B as Backend>::FloatTensorPrimitive<4>

Backward pass for the interpolate operation.

Provided Methods

fn embedding( weights: <B as Backend>::FloatTensorPrimitive<2>, indices: <B as Backend>::IntTensorPrimitive<2>, ) -> <B as Backend>::FloatTensorPrimitive<3>

Embedding operation.

Arguments

• weights - The embedding weights.
• indices - The indices tensor.

Returns

The output tensor.

fn embedding_backward( weights: <B as Backend>::FloatTensorPrimitive<2>, output_grad: <B as Backend>::FloatTensorPrimitive<3>, indices: <B as Backend>::IntTensorPrimitive<2>, ) -> <B as Backend>::FloatTensorPrimitive<2>

Embedding backward operation.

Arguments

• weights - The embedding weights.
• output_grad - The output gradient.
• indices - The indices tensor.

Returns

The gradient.

fn conv1d( x: <B as Backend>::FloatTensorPrimitive<3>, weight: <B as Backend>::FloatTensorPrimitive<3>, bias: Option<<B as Backend>::FloatTensorPrimitive<1>>, options: ConvOptions<1>, ) -> <B as Backend>::FloatTensorPrimitive<3>

One dimensional convolution.

Shapes

x: [batch_size, channels_in, length], weight: [channels_out, channels_in, kernel_size], bias: [channels_out],

fn conv1d_backward( x: <B as Backend>::FloatTensorPrimitive<3>, weight: <B as Backend>::FloatTensorPrimitive<3>, bias: Option<<B as Backend>::FloatTensorPrimitive<1>>, output_grad: <B as Backend>::FloatTensorPrimitive<3>, options: ConvOptions<1>, ) -> Conv1dBackward<B>

Backward pass for the conv1d operation.

fn conv2d_backward( x: <B as Backend>::FloatTensorPrimitive<4>, weight: <B as Backend>::FloatTensorPrimitive<4>, bias: Option<<B as Backend>::FloatTensorPrimitive<1>>, output_grad: <B as Backend>::FloatTensorPrimitive<4>, options: ConvOptions<2>, ) -> Conv2dBackward<B>

Backward pass for the conv2d operation.

fn conv3d_backward( x: <B as Backend>::FloatTensorPrimitive<5>, weight: <B as Backend>::FloatTensorPrimitive<5>, bias: Option<<B as Backend>::FloatTensorPrimitive<1>>, output_grad: <B as Backend>::FloatTensorPrimitive<5>, options: ConvOptions<3>, ) -> Conv3dBackward<B>

Backward pass for the conv3d operation.

fn conv_transpose1d( x: <B as Backend>::FloatTensorPrimitive<3>, weight: <B as Backend>::FloatTensorPrimitive<3>, bias: Option<<B as Backend>::FloatTensorPrimitive<1>>, options: ConvTransposeOptions<1>, ) -> <B as Backend>::FloatTensorPrimitive<3>

One dimensional transposed convolution.

Shapes

x: [batch_size, channels_in, length], weight: [channels_in, channels_out, length], bias: [channels_out],

fn conv_transpose1d_backward( x: <B as Backend>::FloatTensorPrimitive<3>, weight: <B as Backend>::FloatTensorPrimitive<3>, bias: Option<<B as Backend>::FloatTensorPrimitive<1>>, output_grad: <B as Backend>::FloatTensorPrimitive<3>, options: ConvTransposeOptions<1>, ) -> Conv1dBackward<B>

Backward pass for the conv transpose 1d operation.

fn conv_transpose2d_backward( x: <B as Backend>::FloatTensorPrimitive<4>, weight: <B as Backend>::FloatTensorPrimitive<4>, bias: Option<<B as Backend>::FloatTensorPrimitive<1>>, output_grad: <B as Backend>::FloatTensorPrimitive<4>, options: ConvTransposeOptions<2>, ) -> Conv2dBackward<B>

Backward pass for the conv transpose 2d operation.

fn conv_transpose3d_backward( x: <B as Backend>::FloatTensorPrimitive<5>, weight: <B as Backend>::FloatTensorPrimitive<5>, bias: Option<<B as Backend>::FloatTensorPrimitive<1>>, output_grad: <B as Backend>::FloatTensorPrimitive<5>, options: ConvTransposeOptions<3>, ) -> Conv3dBackward<B>

Backward pass for the conv transpose 3d operation.

fn unfold4d( x: <B as Backend>::FloatTensorPrimitive<4>, kernel_size: [usize; 2], options: UnfoldOptions, ) -> <B as Backend>::FloatTensorPrimitive<3>

Four-dimensional unfolding.

Shapes

x: [batch_size, channels_in, height, width], returns: [batch_size, channels_in * kernel_size_1 * kernel_size_2, number of blocks],

fn avg_pool1d( x: <B as Backend>::FloatTensorPrimitive<3>, kernel_size: usize, stride: usize, padding: usize, count_include_pad: bool, ) -> <B as Backend>::FloatTensorPrimitive<3>

One dimensional avg pooling.

Shapes

x: [batch_size, channels, length],

fn avg_pool1d_backward( x: <B as Backend>::FloatTensorPrimitive<3>, grad: <B as Backend>::FloatTensorPrimitive<3>, kernel_size: usize, stride: usize, padding: usize, count_include_pad: bool, ) -> <B as Backend>::FloatTensorPrimitive<3>

Backward pass for the avg pooling 1d operation.

fn adaptive_avg_pool1d( x: <B as Backend>::FloatTensorPrimitive<3>, output_size: usize, ) -> <B as Backend>::FloatTensorPrimitive<3>

One dimensional adaptive avg pooling.

Shapes

x: [batch_size, channels, length],

fn adaptive_avg_pool1d_backward( x: <B as Backend>::FloatTensorPrimitive<3>, grad: <B as Backend>::FloatTensorPrimitive<3>, ) -> <B as Backend>::FloatTensorPrimitive<3>

Backward pass for the adaptive avg pooling 1d operation.

fn max_pool1d( x: <B as Backend>::FloatTensorPrimitive<3>, kernel_size: usize, stride: usize, padding: usize, dilation: usize, ) -> <B as Backend>::FloatTensorPrimitive<3>

One dimensional max pooling.

Shapes

x: [batch_size, channels, length],

fn max_pool1d_with_indices( x: <B as Backend>::FloatTensorPrimitive<3>, kernel_size: usize, stride: usize, padding: usize, dilation: usize, ) -> MaxPool1dWithIndices<B>

One dimensional max pooling with indices.

Shapes

x: [batch_size, channels, height, width],

fn max_pool1d_with_indices_backward( x: <B as Backend>::FloatTensorPrimitive<3>, kernel_size: usize, stride: usize, padding: usize, dilation: usize, output_grad: <B as Backend>::FloatTensorPrimitive<3>, indices: <B as Backend>::IntTensorPrimitive<3>, ) -> MaxPool1dBackward<B>

Backward pass for the max pooling 1d operation.

Object Safety

This trait is not object safe.

Implementations on Foreign Types

impl<B> ModuleOps<Fusion<B>> for Fusion<B>

where B: FusionBackend,

fn conv1d( x: <Fusion<B> as Backend>::FloatTensorPrimitive<3>, weight: <Fusion<B> as Backend>::FloatTensorPrimitive<3>, bias: Option<<Fusion<B> as Backend>::FloatTensorPrimitive<1>>, options: ConvOptions<1>, ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<3>

fn conv2d( x: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, weight: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, bias: Option<<Fusion<B> as Backend>::FloatTensorPrimitive<1>>, options: ConvOptions<2>, ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<4>

fn conv3d( x: <Fusion<B> as Backend>::FloatTensorPrimitive<5>, weight: <Fusion<B> as Backend>::FloatTensorPrimitive<5>, bias: Option<<Fusion<B> as Backend>::FloatTensorPrimitive<1>>, options: ConvOptions<3>, ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<5>

fn conv_transpose1d( x: <Fusion<B> as Backend>::FloatTensorPrimitive<3>, weight: <Fusion<B> as Backend>::FloatTensorPrimitive<3>, bias: Option<<Fusion<B> as Backend>::FloatTensorPrimitive<1>>, options: ConvTransposeOptions<1>, ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<3>

fn conv_transpose2d( x: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, weight: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, bias: Option<<Fusion<B> as Backend>::FloatTensorPrimitive<1>>, options: ConvTransposeOptions<2>, ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<4>

fn conv_transpose3d( x: <Fusion<B> as Backend>::FloatTensorPrimitive<5>, weight: <Fusion<B> as Backend>::FloatTensorPrimitive<5>, bias: Option<<Fusion<B> as Backend>::FloatTensorPrimitive<1>>, options: ConvTransposeOptions<3>, ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<5>

fn avg_pool1d( x: <Fusion<B> as Backend>::FloatTensorPrimitive<3>, kernel_size: usize, stride: usize, padding: usize, count_include_pad: bool, ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<3>

fn avg_pool2d( x: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, kernel_size: [usize; 2], stride: [usize; 2], padding: [usize; 2], count_include_pad: bool, ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<4>

fn avg_pool1d_backward( x: <Fusion<B> as Backend>::FloatTensorPrimitive<3>, grad: <Fusion<B> as Backend>::FloatTensorPrimitive<3>, kernel_size: usize, stride: usize, padding: usize, count_include_pad: bool, ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<3>

fn avg_pool2d_backward( x: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, grad: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, kernel_size: [usize; 2], stride: [usize; 2], padding: [usize; 2], count_include_pad: bool, ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<4>

fn max_pool1d( x: <Fusion<B> as Backend>::FloatTensorPrimitive<3>, kernel_size: usize, stride: usize, padding: usize, dilation: usize, ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<3>

fn max_pool2d( x: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, kernel_size: [usize; 2], stride: [usize; 2], padding: [usize; 2], dilation: [usize; 2], ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<4>

fn max_pool1d_with_indices( x: <Fusion<B> as Backend>::FloatTensorPrimitive<3>, kernel_size: usize, stride: usize, padding: usize, dilation: usize, ) -> MaxPool1dWithIndices<Fusion<B>>

fn max_pool2d_with_indices( x: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, kernel_size: [usize; 2], stride: [usize; 2], padding: [usize; 2], dilation: [usize; 2], ) -> MaxPool2dWithIndices<Fusion<B>>

fn max_pool1d_with_indices_backward( x: <Fusion<B> as Backend>::FloatTensorPrimitive<3>, kernel_size: usize, stride: usize, padding: usize, dilation: usize, output_grad: <Fusion<B> as Backend>::FloatTensorPrimitive<3>, indices: <Fusion<B> as Backend>::IntTensorPrimitive<3>, ) -> MaxPool1dBackward<Fusion<B>>

fn max_pool2d_with_indices_backward( x: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, kernel_size: [usize; 2], stride: [usize; 2], padding: [usize; 2], dilation: [usize; 2], output_grad: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, indices: <Fusion<B> as Backend>::IntTensorPrimitive<4>, ) -> MaxPool2dBackward<Fusion<B>>

fn adaptive_avg_pool1d( x: <Fusion<B> as Backend>::FloatTensorPrimitive<3>, output_size: usize, ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<3>

fn adaptive_avg_pool2d( x: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, output_size: [usize; 2], ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<4>

fn adaptive_avg_pool1d_backward( x: <Fusion<B> as Backend>::FloatTensorPrimitive<3>, grad: <Fusion<B> as Backend>::FloatTensorPrimitive<3>, ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<3>

fn adaptive_avg_pool2d_backward( x: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, grad: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<4>

fn interpolate( x: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, output_size: [usize; 2], options: InterpolateOptions, ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<4>

fn interpolate_backward( x: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, grad: <Fusion<B> as Backend>::FloatTensorPrimitive<4>, output_size: [usize; 2], options: InterpolateOptions, ) -> <Fusion<B> as Backend>::FloatTensorPrimitive<4>

Implementors

impl<B, C> ModuleOps<Autodiff<B, C>> for Autodiff<B, C>

where B: Backend, C: CheckpointStrategy,

impl<E, Q> ModuleOps<LibTorch<E, Q>> for LibTorch<E, Q>

where E: TchElement, Q: QuantElement,

impl<E, Q> ModuleOps<NdArray<E, Q>> for NdArray<E, Q>

where E: FloatNdArrayElement, Q: QuantElement,

impl<F, I> ModuleOps<Candle<F, I>> for Candle<F, I>

where F: FloatCandleElement, I: IntCandleElement,

impl<R, F, I> ModuleOps<JitBackend<R, F, I>> for JitBackend<R, F, I>

where R: JitRuntime, F: FloatElement, I: IntElement,