Adds OUTPUT_PADDING to ConvTrans2D

- Draft state. - Unsure if correct, but a very simple and quick test gives the same result from pytorch. - Note: Tensorflow result differs, both from dfdx and from pytorch. Reference pytorch test: ```python import torch x = np.array([[[[0.1, 0.7], [0.3, 0.4]]]]) w = np.array([[[[-0.1, -0.3, 0.7], [0.8, -0.2, 0.1], [0.3, 0.4, -0.5]]]]) a = torch.nn.ConvTranspose2d(output_padding=0, in_channels=1, out_channels=1, kernel_size=3, stride=2, padding=1, bias = False) b = torch.nn.ConvTranspose2d(output_padding=1, in_channels=1, out_channels=1, kernel_size=3, stride=2, padding=1, bias = False) x = torch.from_numpy(x).float() w0 = torch.from_numpy(w).float() with torch.no_grad(): a.weight = torch.nn.Parameter(w0) b.weight = torch.nn.Parameter(w0) ya = a(x) yb = b(x) print(ya.size()) # torch.Size([1, 1, 3, 3]) print(yb.size()) # torch.Size([1, 1, 4, 4]) print(ya) print(yb) ```
coreylowman · Nov 16, 2023 · 07a6653 · 07a6653
1 parent 4476b5e
commit 07a6653
Show file tree

Hide file tree

Showing 3 changed files with 132 additions and 55 deletions.
diff --git a/dfdx-core/src/tensor_ops/convtrans2d/mod.rs b/dfdx-core/src/tensor_ops/convtrans2d/mod.rs
@@ -51,7 +51,7 @@ pub(super) trait ConvTrans2DKernel<E: Dtype>: Storage<E> {
     ) -> Result<(), Error>;
 }
 
-pub trait TryConvTrans2D<Stride, Padding, Dilation, Groups>: Sized {
+pub trait TryConvTrans2D<Stride, Padding, Dilation, Groups, OutputPadding>: Sized {
     type Convolved;
 
     /// Applies a 2D convolution to the input tensor.
@@ -61,8 +61,9 @@ pub trait TryConvTrans2D<Stride, Padding, Dilation, Groups>: Sized {
         padding: Padding,
         dilation: Dilation,
         groups: Groups,
+        output_padding: OutputPadding,
     ) -> Self::Convolved {
-        self.try_convtrans2d(stride, padding, dilation, groups)
+        self.try_convtrans2d(stride, padding, dilation, groups, output_padding)
             .unwrap()
     }
 
@@ -73,6 +74,7 @@ pub trait TryConvTrans2D<Stride, Padding, Dilation, Groups>: Sized {
         padding: Padding,
         dilation: Dilation,
         groups: Groups,
+        output_padding: OutputPadding,
     ) -> Result<Self::Convolved, Error>;
 }
 
@@ -82,27 +84,31 @@ impl<
         const PADDING: usize,
         const DILATION: usize,
         Groups: Dim,
+        const OUTPUT_PADDING: usize,
         const DIM: usize,
-    > TryConvTrans2D<Const<STRIDE>, Const<PADDING>, Const<DILATION>, Groups>
+    > TryConvTrans2D<Const<STRIDE>, Const<PADDING>, Const<DILATION>, Groups, Const<OUTPUT_PADDING>>
     for (Const<DIM>, Const<KERNEL>)
 where
-    Const<{ (DIM - 1) * STRIDE - 2 * PADDING + DILATION * (KERNEL - 1) + 1 }>: Sized,
+    Const<{ (DIM - 1) * STRIDE - 2 * PADDING + DILATION * (KERNEL - 1) + 1 + OUTPUT_PADDING }>:
+        Sized,
 {
-    type Convolved = Const<{ (DIM - 1) * STRIDE - 2 * PADDING + DILATION * (KERNEL - 1) + 1 }>;
+    type Convolved =
+        Const<{ (DIM - 1) * STRIDE - 2 * PADDING + DILATION * (KERNEL - 1) + 1 + OUTPUT_PADDING }>;
 
     fn try_convtrans2d(
         self,
         _: Const<STRIDE>,
         _: Const<PADDING>,
         _: Const<DILATION>,
         _: Groups,
+        _: Const<OUTPUT_PADDING>,
     ) -> Result<Self::Convolved, Error> {
         Ok(Const)
     }
 }
 
-impl<Kernel: Dim, Stride: Dim, Padding: Dim, Dilation: Dim, Groups: Dim>
-    TryConvTrans2D<Stride, Padding, Dilation, Groups> for (usize, Kernel)
+impl<Kernel: Dim, Stride: Dim, Padding: Dim, Dilation: Dim, Groups: Dim, OutputPadding: Dim>
+    TryConvTrans2D<Stride, Padding, Dilation, Groups, OutputPadding> for (usize, Kernel)
 {
     type Convolved = usize;
 
@@ -112,18 +118,33 @@ impl<Kernel: Dim, Stride: Dim, Padding: Dim, Dilation: Dim, Groups: Dim>
         padding: Padding,
         dilation: Dilation,
         _: Groups,
+        output_padding: OutputPadding,
     ) -> Result<Self::Convolved, Error> {
         let (dim, kernel) = self;
-        Ok(
-            ((dim - 1) * stride.size() + dilation.size() * (kernel.size() - 1) + 1)
-                .checked_sub(2 * padding.size())
-                .unwrap(),
-        )
+        Ok(((dim - 1) * stride.size()
+            + dilation.size() * (kernel.size() - 1)
+            + 1
+            + output_padding.size())
+        .checked_sub(2 * padding.size())
+        .unwrap())
     }
 }
 
-impl<InpChan, OutChanOverGroups, Kernel, Stride, Padding, Dilation, Groups, H, W, E, D, T>
-    TryConvTrans2D<Stride, Padding, Dilation, Groups>
+impl<
+        InpChan,
+        OutChanOverGroups,
+        Kernel,
+        Stride,
+        Padding,
+        Dilation,
+        Groups,
+        OutputPadding,
+        H,
+        W,
+        E,
+        D,
+        T,
+    > TryConvTrans2D<Stride, Padding, Dilation, Groups, OutputPadding>
     for (
         Tensor<(InpChan, H, W), E, D, T>,
         Tensor<(InpChan, OutChanOverGroups, Kernel, Kernel), E, D>,
@@ -136,23 +157,26 @@ where
     Padding: Dim,
     Dilation: Dim,
     Groups: Dim,
+    OutputPadding: Dim,
     H: Dim,
     W: Dim,
     E: Dtype,
     D: ConvTrans2DKernel<E> + crate::tensor_ops::reshape_to::ReshapeKernel<E>,
     T: Tape<E, D>,
     OutChanOverGroups: std::ops::Mul<Groups>,
     <OutChanOverGroups as std::ops::Mul<Groups>>::Output: Dim,
-    (H, Kernel): TryConvTrans2D<Stride, Padding, Dilation, Groups>,
-    (W, Kernel): TryConvTrans2D<Stride, Padding, Dilation, Groups>,
-    <(H, Kernel) as TryConvTrans2D<Stride, Padding, Dilation, Groups>>::Convolved: Dim,
-    <(W, Kernel) as TryConvTrans2D<Stride, Padding, Dilation, Groups>>::Convolved: Dim,
+    (H, Kernel): TryConvTrans2D<Stride, Padding, Dilation, Groups, OutputPadding>,
+    (W, Kernel): TryConvTrans2D<Stride, Padding, Dilation, Groups, OutputPadding>,
+    <(H, Kernel) as TryConvTrans2D<Stride, Padding, Dilation, Groups, OutputPadding>>::Convolved:
+        Dim,
+    <(W, Kernel) as TryConvTrans2D<Stride, Padding, Dilation, Groups, OutputPadding>>::Convolved:
+        Dim,
 {
     type Convolved = Tensor<
         (
             <OutChanOverGroups as std::ops::Mul<Groups>>::Output,
-            <(H, Kernel) as TryConvTrans2D<Stride, Padding, Dilation, Groups>>::Convolved,
-            <(W, Kernel) as TryConvTrans2D<Stride, Padding, Dilation, Groups>>::Convolved,
+            <(H, Kernel) as TryConvTrans2D<Stride, Padding, Dilation, Groups, OutputPadding>>::Convolved,
+            <(W, Kernel) as TryConvTrans2D<Stride, Padding, Dilation, Groups, OutputPadding>>::Convolved,
         ),
         E,
         D,
@@ -165,11 +189,13 @@ where
         padding: Padding,
         dilation: Dilation,
         groups: Groups,
+        output_padding: OutputPadding,
     ) -> Result<Self::Convolved, Error> {
         let (img, filters) = self;
         let (inp_chan, h, w) = img.shape;
         let img = img.try_reshape_like(&(Const::<1>, inp_chan, h, w))?;
-        let out = (img, filters).try_convtrans2d(stride, padding, dilation, groups)?;
+        let out =
+            (img, filters).try_convtrans2d(stride, padding, dilation, groups, output_padding)?;
         let (_, out_chan, out_h, out_w) = out.shape;
         out.try_reshape_like(&(out_chan, out_h, out_w))
     }
@@ -182,13 +208,14 @@ impl<
         Padding,
         Dilation,
         Groups,
+        OutputPadding,
         Batch,
         H,
         W,
         E,
         D,
         T,
-    > TryConvTrans2D<Stride, Padding, Dilation, Groups>
+    > TryConvTrans2D<Stride, Padding, Dilation, Groups, OutputPadding>
     for (
         Tensor<(Batch, InpChan, H, W), E, D, T>,
         Tensor<(InpChan, OutChanOverGroups, Kernel, Kernel), E, D>,
@@ -201,6 +228,7 @@ where
     Padding: Dim,
     Dilation: Dim,
     Groups: Dim,
+    OutputPadding: Dim,
     Batch: Dim,
     H: Dim,
     W: Dim,
@@ -209,17 +237,19 @@ where
     T: Tape<E, D>,
     OutChanOverGroups: std::ops::Mul<Groups>,
     <OutChanOverGroups as std::ops::Mul<Groups>>::Output: Dim,
-    (H, Kernel): TryConvTrans2D<Stride, Padding, Dilation, Groups>,
-    (W, Kernel): TryConvTrans2D<Stride, Padding, Dilation, Groups>,
-    <(H, Kernel) as TryConvTrans2D<Stride, Padding, Dilation, Groups>>::Convolved: Dim,
-    <(W, Kernel) as TryConvTrans2D<Stride, Padding, Dilation, Groups>>::Convolved: Dim,
+    (H, Kernel): TryConvTrans2D<Stride, Padding, Dilation, Groups, OutputPadding>,
+    (W, Kernel): TryConvTrans2D<Stride, Padding, Dilation, Groups, OutputPadding>,
+    <(H, Kernel) as TryConvTrans2D<Stride, Padding, Dilation, Groups, OutputPadding>>::Convolved:
+        Dim,
+    <(W, Kernel) as TryConvTrans2D<Stride, Padding, Dilation, Groups, OutputPadding>>::Convolved:
+        Dim,
 {
     type Convolved = Tensor<
         (
             Batch,
             <OutChanOverGroups as std::ops::Mul<Groups>>::Output,
-            <(H, Kernel) as TryConvTrans2D<Stride, Padding, Dilation, Groups>>::Convolved,
-            <(W, Kernel) as TryConvTrans2D<Stride, Padding, Dilation, Groups>>::Convolved,
+            <(H, Kernel) as TryConvTrans2D<Stride, Padding, Dilation, Groups, OutputPadding>>::Convolved,
+            <(W, Kernel) as TryConvTrans2D<Stride, Padding, Dilation, Groups, OutputPadding>>::Convolved,
         ),
         E,
         D,
@@ -232,6 +262,7 @@ where
         padding: Padding,
         dilation: Dilation,
         groups: Groups,
+        output_padding: OutputPadding,
     ) -> Result<Self::Convolved, Error> {
         let (img, filters) = self;
         assert_eq!(img.shape.1, filters.shape.0);
@@ -242,8 +273,8 @@ where
         if img.strides != img.shape.strides() || filters.strides != filters.shape.strides() {
             panic!("Image & filter inputs to conv2d must be contiguous");
         }
-        let h_out = (h, kernel).convtrans2d(stride, padding, dilation, groups);
-        let w_out = (w, kernel).convtrans2d(stride, padding, dilation, groups);
+        let h_out = (h, kernel).convtrans2d(stride, padding, dilation, groups, output_padding);
+        let w_out = (w, kernel).convtrans2d(stride, padding, dilation, groups, output_padding);
         let op = ConvTrans2DOp {
             stride: stride.size(),
             padding: padding.size(),

diff --git a/dfdx-core/src/tensor_ops/convtrans2d/tests.rs b/dfdx-core/src/tensor_ops/convtrans2d/tests.rs
@@ -33,8 +33,8 @@ fn test_convtrans2d_default() {
             ],
         ])
         .to_dtype::<TestDtype>();
-    let y =
-        (x.leaky_trace(), w.clone()).convtrans2d(Const::<1>, Const::<0>, Const::<1>, Const::<1>);
+    let y = (x.leaky_trace(), w.clone())
+        .convtrans2d(Const::<1>, Const::<0>, Const::<1>, Const::<1>, Const::<0>);
     #[rustfmt::skip]
     assert_close_to_literal!(
         y,
@@ -125,8 +125,8 @@ fn test_convtrans2d_stride_2() {
             ],
         ])
         .to_dtype::<TestDtype>();
-    let y =
-        (x.leaky_trace(), w.clone()).convtrans2d(Const::<2>, Const::<0>, Const::<1>, Const::<1>);
+    let y = (x.leaky_trace(), w.clone())
+        .convtrans2d(Const::<2>, Const::<0>, Const::<1>, Const::<1>, Const::<0>);
     #[rustfmt::skip]
     assert_close_to_literal!(
         y,
@@ -223,8 +223,8 @@ fn test_convtrans2d_padded() {
             ],
         ])
         .to_dtype::<TestDtype>();
-    let y =
-        (x.leaky_trace(), w.clone()).convtrans2d(Const::<1>, Const::<1>, Const::<1>, Const::<1>);
+    let y = (x.leaky_trace(), w.clone())
+        .convtrans2d(Const::<1>, Const::<1>, Const::<1>, Const::<1>, Const::<0>);
     assert_close_to_literal!(
         y,
         [
@@ -283,8 +283,8 @@ fn test_convtrans2d_batched() {
     let x: Tensor<Rank3<3, 28, 28>, TestDtype, _> = dev.sample_normal();
     let w: Tensor<Rank4<3, 5, 6, 6>, TestDtype, _> = dev.sample_normal();
 
-    let y: Tensor<Rank3<5, 83, 83>, _, _, _> =
-        (x.leaky_trace(), w.clone()).convtrans2d(Const::<3>, Const::<2>, Const::<1>, Const::<1>);
+    let y: Tensor<Rank3<5, 83, 83>, _, _, _> = (x.leaky_trace(), w.clone())
+        .convtrans2d(Const::<3>, Const::<2>, Const::<1>, Const::<1>, Const::<0>);
     let y0 = y.retaped::<NoneTape>();
     let grads0 = y.square().mean().backward();
     let x0 = grads0.get(&x);
@@ -294,8 +294,8 @@ fn test_convtrans2d_batched() {
         .broadcast::<Rank4<10, 3, 28, 28>, _>()
         .reshape::<Rank4<10, 3, 28, 28>>();
 
-    let y: Tensor<Rank4<10, 5, 83, 83>, _, _, _> =
-        (x.leaky_trace(), w.clone()).convtrans2d(Const::<3>, Const::<2>, Const::<1>, Const::<1>);
+    let y: Tensor<Rank4<10, 5, 83, 83>, _, _, _> = (x.leaky_trace(), w.clone())
+        .convtrans2d(Const::<3>, Const::<2>, Const::<1>, Const::<1>, Const::<0>);
     for i in 0..10 {
         assert_close_to_tensor!(y0, y.retaped::<NoneTape>().select(dev.tensor(i)), 1e-5);
     }
@@ -341,8 +341,8 @@ fn test_convtrans2d_grouped() {
             ],
         ])
         .to_dtype::<TestDtype>();
-    let y =
-        (x.leaky_trace(), w.clone()).convtrans2d(Const::<1>, Const::<0>, Const::<1>, Const::<2>);
+    let y = (x.leaky_trace(), w.clone())
+        .convtrans2d(Const::<1>, Const::<0>, Const::<1>, Const::<2>, Const::<0>);
     #[rustfmt::skip]
     assert_close_to_literal!(
         y,
@@ -451,8 +451,8 @@ fn test_convtrans2d_dilated() {
             ],
         ])
         .to_dtype::<TestDtype>();
-    let y =
-        (x.leaky_trace(), w.clone()).convtrans2d(Const::<1>, Const::<0>, Const::<2>, Const::<1>);
+    let y = (x.leaky_trace(), w.clone())
+        .convtrans2d(Const::<1>, Const::<0>, Const::<2>, Const::<1>, Const::<0>);
     #[rustfmt::skip]
     assert_close_to_literal!(
         y,