2D Convolution

group 2D Convolution

Functions

NVCVStatus cvcudaConv2DCreate(NVCVOperatorHandle *handle)

Constructs an instance of the Conv2D.

Parameters:

handle – [out] Where the operator instance handle will be written to.

  • Must not be NULL.

Return values:

  • NVCV_ERROR_INVALID_ARGUMENT – Handle is null.

  • NVCV_ERROR_OUT_OF_MEMORY – Not enough memory to create the operator.

  • NVCV_SUCCESS – Operation executed successfully.

NVCVStatus cvcudaConv2DVarShapeSubmit(NVCVOperatorHandle handle, cudaStream_t stream, NVCVImageBatchHandle in, NVCVImageBatchHandle out, NVCVImageBatchHandle kernel, NVCVTensorHandle kernelAnchor, NVCVBorderType borderMode)

Executes the Conv2D operation on the given cuda stream. This operation does not wait for completion.

Limitations:

Input: Data Layout: [kNHWC, kHWC, kNCHW, kCHW] Channels: 1, 3, 4

Data Type

Allowed

8bit Unsigned

Yes

8bit Signed

No

16bit Unsigned

Yes

16bit Signed

Yes

32bit Unsigned

No

32bit Signed

Yes

16bit Float

No

32bit Float

Yes

64bit Float

No

Output: Data Layout: [kNHWC, kHWC, kNCHW, kCHW] Channels: 1, 3, 4

Data Type

Allowed

8bit Unsigned

Yes

8bit Signed

No

16bit Unsigned

Yes

16bit Signed

Yes

32bit Unsigned

No

32bit Signed

Yes

16bit Float

No

32bit Float

Yes

64bit Float

No

Input/Output dependency 

 Property      |  Input == Output
-------------- | -------------
 Data Layout   | Yes
 Data Type     | Yes
 Number        | Yes
 Channels      | Yes
 Width         | Yes
 Height        | Yes

Parameters:

  • handle – [in] Handle to the operator.

    • Must not be NULL.

  • stream – [in] Handle to a valid CUDA stream.

  • in – [in] Input tensor.

  • out – [out] Output tensor.

  • kernel – [in] Convolution kernels (one for each batch image) to be used. Each image width and height correspond to the kernel width and height.

    • Must be of pixel type NVCV_DATA_TYPE_F32

  • kernelAnchor – [in] 1D Tensor with the anchor of each kernel (one for each batch image). The anchor (x, y) indicates the relative position of a filtered point within the kernel. (-1, -1) means that the anchor is at the kernel center.

    • Must be of pixel type NVCV_DATA_TYPE_2S32

  • borderMode – [in] Border mode to be used when accessing elements outside, cf. NVCVBorderType.

Return values:

  • NVCV_ERROR_INVALID_ARGUMENT – Some parameter is outside valid range.

  • NVCV_ERROR_INTERNAL – Internal error in the operator, invalid types passed in.

  • NVCV_SUCCESS – Operation executed successfully.