Audio PR 8 - Resample + Operator Overloading Nodes #34

SundarRajan28 · 2024-03-18T15:08:54Z

No description provided.

sampath1117

Please address review comments

sampath1117 · 2024-03-26T15:43:19Z

amd_openvx_extensions/amd_rpp/source/kernel_rpp.cpp

+    return node;
+}
+
+VX_API_ENTRY vx_node VX_API_CALL vxExtRppTensorMulScalar(vx_graph graph, vx_tensor pSrc, vx_tensor pDst, vx_scalar scalar_value, vx_uint32 nbatchSize) {


is it necessary to pass batchsize as a param to this function?
for other functions, i have not seen batchsize being passed as a separate argument

nBatchSize is not used in this function, so I think it is safe to remove

sampath1117 · 2024-03-26T15:44:28Z

amd_openvx_extensions/amd_rpp/source/kernel_rpp.cpp

+    vx_node node = NULL;
+    vx_context context = vxGetContext((vx_reference)graph);
+    if (vxGetStatus((vx_reference)context) == VX_SUCCESS) {
+        vx_uint32 dev_type = getGraphAffinity(graph);


Please change dev_type to deviceType

sampath1117 · 2024-03-26T15:47:13Z

amd_openvx_extensions/amd_rpp/source/kernel_rpp.cpp

+    vx_node node = NULL;
+    vx_context context = vxGetContext((vx_reference)graph);
+    if (vxGetStatus((vx_reference)context) == VX_SUCCESS) {
+        vx_uint32 dev_type = getGraphAffinity(graph);


same comment. change dev_type to deviceType

sampath1117 · 2024-03-26T15:51:10Z

amd_openvx_extensions/amd_rpp/source/tensor/TensorAddTensor.cpp

+        data->pDstRoi[i].xywhROI.xy.x = data->pSrcRoi[i].xywhROI.xy.x;
+        data->pDstRoi[i].xywhROI.xy.y = data->pSrcRoi[i].xywhROI.xy.y;
+    }
+}


I think this might be need not here. Remember this has been removed with preemphasis too
Please check with @swetha097

Yes, here we are just copying src roi to dst roi so we need not do this @SundarRajan28 please check

So please do not pass the pDstRoi to this node.

We are setting the dstROI values from srcROI in this function. @swetha097 should this be removed?

sampath1117 · 2024-03-26T15:52:28Z

amd_openvx_extensions/amd_rpp/source/tensor/TensorAddTensor.cpp

+    return status;
+}
+
+static vx_status VX_CALLBACK validateTensorAddTensor(vx_node node, const vx_reference parameters[], vx_uint32 num, vx_meta_format metas[]) {


Please remove this "num" variable if not used
I am seeing unused in most of the audio PR's. Please check and remove this if not needed

Its used in all openvx extensions so this may be a syntax for the openvx extensions

sampath1117 · 2024-03-26T15:55:53Z

amd_openvx_extensions/amd_rpp/source/tensor/TensorMulScalar.cpp

+//! \brief The kernel target support callback.
+// TODO::currently the node is setting the same affinity as context. This needs to change when we have hubrid modes in the same graph
+static vx_status VX_CALLBACK query_target_support(vx_graph graph, vx_node node,
+                                                  vx_bool use_opencl_1_2,               // [input]  false: OpenCL driver is 2.0+; true: OpenCL driver is 1.2


Please remove the OpenCL related comments

sampath1117 · 2024-03-26T15:56:42Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+#include "internal_publishKernels.h"
+#include "vx_ext_amd.h"
+#define NUM_OF_DIMS 5
+


is vx_ext_amd.h and NUM_OF_DIMS needed to be declared here?

sampath1117 · 2024-03-26T16:01:07Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+    cl_mem cl_pDst;
+#elif ENABLE_HIP
+    void *hip_pSrc;
+    void *hip_pDst;


I think we don't require HIP and OpenCL related variables in new changes done in all openvx nodes
I am not able to see this variables in brightness.cpp
https://github.com/swetha097/MIVisionX/blob/swbs_m5/audio/pr8/amd_openvx_extensions/amd_rpp/source/tensor/Brightness.cpp

Please check and remove this

sampath1117 · 2024-03-26T16:01:59Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+    RppPtr_t pDst;
+    float quality;
+    uint maxDstWidth;
+    RpptDescPtr src_desc_ptr;


Please change variables from snake case to camel case.
Please check brightness and make the required changes
https://github.com/swetha097/MIVisionX/blob/swbs_m5/audio/pr8/amd_openvx_extensions/amd_rpp/source/tensor/Brightness.cpp

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

fiona-gladwin · 2024-03-27T16:53:38Z

amd_openvx_extensions/amd_rpp/CMakeLists.txt

@@ -161,6 +161,9 @@ list(APPEND SOURCES
        source/tensor/WarpAffine.cpp
        source/tensor/SequenceRearrange.cpp
        source/tensor/PreemphasisFilter.cpp
+        source/tensor/Resample.cpp


Add in Alphabetical order itself

fiona-gladwin · 2024-03-27T16:56:44Z

amd_openvx_extensions/amd_rpp/include/kernels_rpp.h

+        VX_KERNEL_RPP_TODECIBELS = VX_KERNEL_BASE(VX_ID_AMD, VX_LIBRARY_RPP) + 0x77,
+        VX_KERNEL_RPP_RESAMPLE = VX_KERNEL_BASE(VX_ID_AMD, VX_LIBRARY_RPP) + 0x78,
+        VX_KERNEL_RPP_TENSORMULSCALAR = VX_KERNEL_BASE(VX_ID_AMD, VX_LIBRARY_RPP) + 0x79,
+        VX_KERNEL_RPP_TENSORADDTENSOR = VX_KERNEL_BASE(VX_ID_AMD, VX_LIBRARY_RPP) + 0x80


After 79 it is 7a

fiona-gladwin · 2024-03-27T16:58:58Z

amd_openvx_extensions/amd_rpp/include/vx_ext_rpp.h

@@ -1919,6 +1919,55 @@ extern "C"
 	 */
 	SHARED_PUBLIC vx_node VX_API_CALL vxExtRppToDecibels(vx_graph graph, vx_tensor pSrc, vx_tensor pSrcRoi, vx_tensor pDst, vx_scalar cutOffDB, vx_scalar multiplier, vx_scalar referenceMagnitude);

+	/*! \brief [Graph] Creates a RPP Resample function node.


Can we change it to better description like, Produces a resampled output

Check other functions

fiona-gladwin · 2024-03-27T16:59:33Z

amd_openvx_extensions/amd_rpp/include/vx_ext_rpp.h

+	/*! \brief [Graph] Creates a RPP Resample function node.
+	 * \ingroup group_amd_rpp
+	 * \param [in] graph The handle to the graph.
+	 * \param [in] pSrc The input tensor in <tt>\ref VX_TYPE_UINT8<tt> or <tt>\ref VX_TYPE_FLOAT32<tt> or


Please check the supported data types and correct it for pSrc and pDst

fiona-gladwin · 2024-03-27T17:01:07Z

amd_openvx_extensions/amd_rpp/include/vx_ext_rpp.h

+	 * <tt>\ref VX_TYPE_FLOAT16<tt> or <tt>\ref VX_TYPE_INT8<tt> format data.
+	 * \param [out] pDst The output tensor in <tt>\ref VX_TYPE_UINT8<tt> or <tt>\ref VX_TYPE_FLOAT32<tt> or
+	 * <tt>\ref VX_TYPE_FLOAT16<tt> or <tt>\ref VX_TYPE_INT8<tt> format data.
+	 * \param [in] pSrcDims The input tensor of batch size in <tt>unsigned int<tt> containing the roi values for the input in xywh/ltrb format.


If it contains ROI rename it as pSrcRoi and pDstROI, Also remove the xywh/ltrb format

Please remove from Spectrogram also it is not changed

fiona-gladwin · 2024-03-27T17:43:15Z

amd_openvx_extensions/amd_rpp/source/tensor/TensorMulScalar.cpp

+    size_t tensor_dims[RPP_MAX_TENSOR_DIMS];
+    vx_enum tensor_type;
+    STATUS_ERROR_CHECK(vxQueryTensor((vx_tensor)parameters[1], VX_TENSOR_NUMBER_OF_DIMS, &num_tensor_dims, sizeof(num_tensor_dims)));
+    if (num_tensor_dims < 3) return ERRMSG(VX_ERROR_INVALID_DIMENSION, "validate TensorMulScalar: tensor #0 dimensions=%lu (must be greater than or equal to 3)\n", num_tensor_dims);


Also it is tensor #1 please change

fiona-gladwin · 2024-03-27T17:43:30Z

amd_openvx_extensions/amd_rpp/source/tensor/TensorMulScalar.cpp

+    STATUS_ERROR_CHECK(vxQueryNode(node, VX_NODE_LOCAL_DATA_PTR, &data, sizeof(data)));
+    refreshTensorMulScalar(node, parameters, num, data);
+    if (data->deviceType == AGO_TARGET_AFFINITY_GPU) {
+#if ENABLE_OPENCL


Combine OCL and HIP

fiona-gladwin · 2024-03-27T17:43:44Z

amd_openvx_extensions/amd_rpp/source/tensor/TensorMulScalar.cpp

+        status = VX_ERROR_NOT_IMPLEMENTED;
+#endif
+    } else if (data->deviceType == AGO_TARGET_AFFINITY_CPU) {
+        if (data->inputTensorType == vx_type_e::VX_TYPE_FLOAT32 && data->outputTensorType == vx_type_e::VX_TYPE_FLOAT32) {


else return some unsupported status

fiona-gladwin · 2024-03-27T17:44:16Z

amd_openvx_extensions/amd_rpp/source/tensor/TensorMulScalar.cpp

+    data->tensorSize = 1;
+    for (int i = 0; i < num_of_dims; i++)
+        data->tensorSize *= tensor_dims[i];
+    if (data->inputTensorType == vx_type_e::VX_TYPE_FLOAT32 && data->outputTensorType == vx_type_e::VX_TYPE_FLOAT32)


else throw some error, or return unsupported status

fiona-gladwin · 2024-03-27T17:45:47Z

amd_openvx_extensions/amd_rpp/source/tensor/TensorMulScalar.cpp

+}
+
+//! \brief The kernel target support callback.
+// TODO::currently the node is setting the same affinity as context. This needs to change when we have hubrid modes in the same graph


Change spelling of hubrid to hybrid, applicable to other new files as well

sampath1117

Please address review comments for Resample OpenVX

sampath1117 · 2024-04-01T12:35:46Z

amd_openvx_extensions/amd_rpp/CMakeLists.txt

        source/tensor/ToDecibels.cpp
        source/tensor/Vignette.cpp
        source/tensor/WarpAffine.cpp
-        source/tensor/SequenceRearrange.cpp


Why sequence rearrange is deleted?

It is present twice

sampath1117 · 2024-04-01T12:39:37Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+    RpptDescPtr pSrcDesc;
+    RpptDescPtr pDstDesc;
+    Rpp32s *pSampleFrames;
+    Rpp32s *pSampleChannels;


Frames and channels are combined into single RPP Tensor in latest resample kernel
Please combine these in rocAL too
https://github.com/ROCm/rpp/blob/ea7c9f14d8773a2c1b359b679869403e26789438/src/modules/cpu/kernel/resample.hpp#L35

Is this addressed?

sampath1117 · 2024-04-01T12:42:20Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+        if (data->pInRateTensor[i] != 0)
+            scale_ratio = data->pOutRateTensor[i] / static_cast<float>(data->pInRateTensor[i]);
+        else
+            scale_ratio = 0;


Use it as a ternary operator here

scale_ratio = (data->pInRateTensor[i] != 0) ? (data->pOutRateTensor[i] / static_cast<float>(data->pInRateTensor[i])) : 0;

Ideally if scale_ratio is 0, we need to throw an error

sampath1117 · 2024-04-01T12:42:41Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+            scale_ratio = data->pOutRateTensor[i] / static_cast<float>(data->pInRateTensor[i]);
+        else
+            scale_ratio = 0;
+        dst_roi[i].xywhROI.roiWidth = (int)std::ceil(scale_ratio * src_roi[i].xywhROI.roiWidth);


Please use a static cast instead of c style casting

Also since this is Audio, inferring the ROI as width and height does not seem right to me, Please check Sundar

We are using .roiWidth and .roiHeight in all the audio openvx extensions

sampath1117 · 2024-04-01T12:44:09Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+        else
+            scale_ratio = 0;
+        dst_roi[i].xywhROI.roiWidth = (int)std::ceil(scale_ratio * src_roi[i].xywhROI.roiWidth);
+        dst_roi[i].xywhROI.roiHeight = src_roi[i].xywhROI.roiHeight;


Height should be interpreted as samples and width should be interpreted as channels
So this should be ideally changed

Please check with @swetha097 for further clarifications

The variables are named like this in RpptROI struct

sampath1117 · 2024-04-01T12:45:18Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+#if ENABLE_OPENCL
+        return VX_ERROR_NOT_IMPLEMENTED;
+#elif ENABLE_HIP
+        return VX_ERROR_NOT_IMPLEMENTED;


Combine HIP and OpenCL and return single error

sampath1117 · 2024-04-01T12:46:11Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+    }
+    if (data->deviceType == AGO_TARGET_AFFINITY_CPU) {
+        rpp_status = rppt_resample_host(data->pSrc, data->pSrcDesc, data->pDst, data->pDstDesc,
+                                        data->pInRateTensor, data->pOutRateTensor, data->pSampleFrames, data->window, data->handle->rppHandle);


You need to combine frames and channels and pass a single tensor
This code will not work, since only frames is passed here

sampath1117 · 2024-04-01T12:47:08Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+    int lobes = std::round(0.007 * data->quality * data->quality - 0.09 * data->quality + 3);
+    int lookupSize = lobes * 64 + 1;
+    windowed_sinc(data->window, lookupSize, lobes);
+    refreshResample(node, parameters, num, data);


Please check and remove this refreshResample since in process, refreshResample is called anyway

sampath1117 · 2024-04-01T12:48:30Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+    data->pDstDesc->offsetInBytes = 0;
+    fillAudioDescriptionPtrFromDims(data->pDstDesc, data->outputTensorDims);
+
+    data->pSampleFrames = static_cast<signed int *>(calloc(data->pSrcDesc->n, sizeof(unsigned int)));


Incase of HIP backend
Would suggest to have this buffers in pinned memory, which is what might be needed

Please check with @snehaa8

fiona-gladwin

Please check the comments and add changes

fiona-gladwin · 2024-04-01T13:19:23Z

amd_openvx_extensions/amd_rpp/CMakeLists.txt

@@ -147,7 +147,9 @@ list(APPEND SOURCES
        source/tensor/Nop.cpp
        source/tensor/Rotate.cpp
        source/tensor/Pixelate.cpp
+        source/tensor/PreemphasisFilter.cpp


Can you add this to PreEmphasis filter PR?

fiona-gladwin · 2024-04-01T13:19:36Z

amd_openvx_extensions/amd_rpp/CMakeLists.txt

        source/tensor/ToDecibels.cpp
        source/tensor/Vignette.cpp
        source/tensor/WarpAffine.cpp
-        source/tensor/SequenceRearrange.cpp


It is present twice

fiona-gladwin · 2024-04-01T14:54:48Z

amd_openvx_extensions/amd_rpp/source/kernel_rpp.cpp

+            (vx_reference)pSrc1,
+            (vx_reference)pSrc2,
+            (vx_reference)pDst,
+            (vx_reference)srcRoi,


Change to pSrcRoi and pDstRoi

fiona-gladwin · 2024-04-01T14:55:55Z

amd_openvx_extensions/amd_rpp/include/vx_ext_rpp.h

+	 * \param [in] pSrc2 The input tensor2 in <tt>\ref VX_TYPE_FLOAT32<tt> format data.
+	 * \param [out] pDst The output tensor in <tt>\ref VX_TYPE_FLOAT32<tt> format data.
+	 * \param [in] pSrcRoi The input tensor of batch size in <tt>unsigned int<tt> containing the roi values for the input1.
+	 * \param [in] pDstRoi The input tensor of batch size in <tt>unsigned int<tt> containing the roi values for the input2.


Is it the ROI for tensor 2?

Please check it and rename accordingly

This is the ROI for the output tensor

fiona-gladwin · 2024-04-01T14:56:52Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+    RpptDescPtr pSrcDesc;
+    RpptDescPtr pDstDesc;
+    Rpp32s *pSampleFrames;
+    Rpp32s *pSampleChannels;


Is this addressed?

fiona-gladwin · 2024-04-01T15:05:18Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+    if (data->pSampleFrames != nullptr) free(data->pSampleFrames);
+    if (data->pSampleChannels != nullptr) free(data->pSampleChannels);
+    if (data->pInRateTensor != nullptr) free(data->pInRateTensor);
+    delete (data->pSrcDesc);


Remove brackets

fiona-gladwin · 2024-04-01T15:05:28Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+    ResampleLocalData *data;
+    STATUS_ERROR_CHECK(vxQueryNode(node, VX_NODE_LOCAL_DATA_PTR, &data, sizeof(data)));
+    STATUS_ERROR_CHECK(releaseRPPHandle(node, data->handle, data->deviceType));
+    if (data->pSampleFrames != nullptr) free(data->pSampleFrames);


Use only delete and new operator

fiona-gladwin · 2024-04-01T15:05:42Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+    fillAudioDescriptionPtrFromDims(data->pDstDesc, data->outputTensorDims);
+
+    data->pSampleFrames = static_cast<signed int *>(calloc(data->pSrcDesc->n, sizeof(unsigned int)));
+    data->pSampleChannels = static_cast<signed int *>(calloc(data->pSrcDesc->n, sizeof(unsigned int)));


Use new operator instead of calloc

fiona-gladwin · 2024-04-01T15:07:08Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+            scale_ratio = data->pOutRateTensor[i] / static_cast<float>(data->pInRateTensor[i]);
+        else
+            scale_ratio = 0;
+        dst_roi[i].xywhROI.roiWidth = (int)std::ceil(scale_ratio * src_roi[i].xywhROI.roiWidth);


Also since this is Audio, inferring the ROI as width and height does not seem right to me, Please check Sundar

fiona-gladwin · 2024-04-01T15:07:38Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

Please address all review comments from Sampath

sampath1117

Please address review comments

sampath1117 · 2024-05-14T07:00:47Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+        window.lookup[i + 1] = w;
+    }
+    window.center = center + 1;
+    window.scale = 1 / scale;


Please change 1/scale as 1.0f/scale as per rajys comment

sampath1117 · 2024-05-14T07:01:47Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+
+static vx_status VX_CALLBACK refreshResample(vx_node node, const vx_reference *parameters, vx_uint32 num, ResampleLocalData *data) {
+    vx_status status = VX_SUCCESS;
+    int nDim = 2;  // Num dimensions for audio tensor


why nDim =2 is hardcoded here?

sampath1117 · 2024-05-14T07:02:27Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+        STATUS_ERROR_CHECK(vxQueryTensor((vx_tensor)parameters[2], VX_TENSOR_BUFFER_HOST, &roi_tensor_ptr_src, sizeof(roi_tensor_ptr_src)));
+        STATUS_ERROR_CHECK(vxQueryTensor((vx_tensor)parameters[3], VX_TENSOR_BUFFER_HOST, &roi_tensor_ptr_dst, sizeof(roi_tensor_ptr_dst)));
+        if (!data->pSrcRoi) {
+            data->pSrcRoi = new Rpp32s[data->pSrcDesc->n * nDim];


This has to be done in initialize not refresh?

sampath1117 · 2024-05-14T07:04:01Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+    RpptROI *src_roi = reinterpret_cast<RpptROI *>(roi_tensor_ptr_src);
+    RpptROI *dst_roi = reinterpret_cast<RpptROI *>(roi_tensor_ptr_dst);
+    update_destination_roi(data, src_roi, dst_roi);
+    for (unsigned i = 0; i < data->pSrcDesc->n; i++) {


Format the loop as per comment in downmixing pr
#32 (comment)

sampath1117 · 2024-05-14T07:04:46Z

amd_openvx_extensions/amd_rpp/source/tensor/Resample.cpp

+    if (data->pSrcDesc) delete data->pSrcDesc;
+    if (data->pDstDesc) delete data->pDstDesc;
+    STATUS_ERROR_CHECK(releaseRPPHandle(node, data->handle, data->deviceType));
+    if (data) delete data;


Please check the validity of data also before delete

sampath1117 · 2024-05-14T07:07:08Z

amd_openvx_extensions/amd_rpp/source/tensor/TensorAddTensor.cpp

+        STATUS_ERROR_CHECK(vxQueryTensor((vx_tensor)parameters[1], VX_TENSOR_DATA_TYPE, &data->inTensorType, sizeof(data->inTensorType)));
+        STATUS_ERROR_CHECK(vxQueryTensor((vx_tensor)parameters[2], VX_TENSOR_DATA_TYPE, &data->outTensorType, sizeof(data->outTensorType)));
+
+        refreshTensorAddTensor(node, parameters, num, data);


Please remove this refresh since this should be called in process anyway

sampath1117 · 2024-05-14T07:07:45Z

amd_openvx_extensions/amd_rpp/source/tensor/TensorMulScalar.cpp

+    Rpp32u deviceType;
+    RppPtr_t pSrc;
+    RppPtr_t pDst;
+    float scalarValueFloat;


remove Float from variable name
just have it as scalarValue

sampath1117 · 2024-05-14T07:08:31Z

amd_openvx_extensions/amd_rpp/source/tensor/TensorMulScalar.cpp

+    size_t num_tensor_dims;
+    STATUS_ERROR_CHECK(vxQueryTensor((vx_tensor)parameters[0], VX_TENSOR_NUMBER_OF_DIMS, &num_tensor_dims, sizeof(num_tensor_dims)));
+
+    // Validate for output parameters


Are there no validation of num_tensor_dims needed here?

sampath1117 · 2024-05-14T07:09:07Z

amd_openvx_extensions/amd_rpp/source/tensor/TensorMulScalar.cpp

+            float *srcPtrTemp = static_cast<float *>(data->pSrc);
+            float *dstPtrTemp = static_cast<float *>(data->pDst);
+            uint bufferLength = data->tensorSize / sizeof(float);
+            uint alignedLength = (bufferLength / 8) * 8;


Please address this comment
https://github.com/swetha097/MIVisionX/pull/34/files#r1546423884

sampath1117 · 2024-05-14T07:09:59Z

amd_openvx_extensions/amd_rpp/source/tensor/TensorMulScalar.cpp

+            data->tensorSize *= sizeof(float);
+        else
+            return VX_ERROR_NOT_SUPPORTED;
+        refreshTensorMulScalar(node, parameters, num, data);


Please remove this refresh since anyway called in process node

* Bump rocm-docs-core[api_reference] from 0.34.0 to 0.34.2 in /docs/sphinx (ROCm#1286) Bumps [rocm-docs-core[api_reference]](https://github.com/RadeonOpenCompute/rocm-docs-core) from 0.34.0 to 0.34.2. - [Release notes](https://github.com/RadeonOpenCompute/rocm-docs-core/releases) - [Changelog](https://github.com/ROCm/rocm-docs-core/blob/develop/CHANGELOG.md) - [Commits](ROCm/rocm-docs-core@v0.34.0...v0.34.2) --- updated-dependencies: - dependency-name: rocm-docs-core[api_reference] dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * Bump cryptography from 42.0.0 to 42.0.2 in /docs/sphinx (ROCm#1289) Bumps [cryptography](https://github.com/pyca/cryptography) from 42.0.0 to 42.0.2. - [Changelog](https://github.com/pyca/cryptography/blob/main/CHANGELOG.rst) - [Commits](pyca/cryptography@42.0.0...42.0.2) --- updated-dependencies: - dependency-name: cryptography dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * Add PreEmphasis filter support * Fix ROI - change from xy to width & height * Adding openvx changes for downmix node * Audio Augmentations 1 PR - NSR and Spectrogram * Revert "Adding openvx changes for downmix node" This reverts commit d53f81d. * Merging swbs_m2/audio/pr5 into swbs_m3/audio/pr6 * Fixing build issues * PR comments resolution in the PreEmphais Filter * Minor Changes * Change the borderType enum to int32 from uint32 dtype * Fix validation of preemphasis * Remove the memcopy of the src and dest rois as it can be handled in the rocAL - since the src and dst rois remain same * Formatting change - minor * Remove NSR * Minor formatting changes * Minor fix * Minor update - remove the 2nd instance of preemphasis filter * Add openvx extensions for to_decibels augmentations * Enum dtype - change from uint to int * Remove roi_tensor_ptr_dst as its unused after latest changes * Remove the dst_roi arg from vxExtRppPreemphasisFilter call as its unused * Add MFB to MIVisisonX * Revert "Add MFB to MIVisisonX" This reverts commit dc4200b. * Fixing issues with src ROI * Resolve the PR comments * Resolving review comments * Resolving review comments * Change the dims[0] and dims[1] positioning for Spectrogram and AudioFillDescPointers * Removing dstROI from to_decibels openvx augmentation * Change function name to camelCase * Resolving review comments * Revert "Change the dims[0] and dims[1] positioning for Spectrogram and AudioFillDescPointers" This reverts commit 886d6af. * Fix Spectrogram * Docs - update TOC for API Ref (ROCm#1327) * Bump rocm-docs-core[api_reference] from 0.38.0 to 0.38.1 in /docs/sphinx (ROCm#1328) Bumps [rocm-docs-core[api_reference]](https://github.com/RadeonOpenCompute/rocm-docs-core) from 0.38.0 to 0.38.1. - [Release notes](https://github.com/RadeonOpenCompute/rocm-docs-core/releases) - [Changelog](https://github.com/ROCm/rocm-docs-core/blob/develop/CHANGELOG.md) - [Commits](ROCm/rocm-docs-core@v0.38.0...v0.38.1) --- updated-dependencies: - dependency-name: rocm-docs-core[api_reference] dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * Update CHANGELOG.md * Update CHANGELOG.md * Documents - Bump idna from 3.4 to 3.7 in /docs/sphinx (ROCm#1330) Bumps [idna](https://github.com/kjd/idna) from 3.4 to 3.7. - [Release notes](https://github.com/kjd/idna/releases) - [Changelog](https://github.com/kjd/idna/blob/master/HISTORY.rst) - [Commits](kjd/idna@v3.4...v3.7) --- updated-dependencies: - dependency-name: idna dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * Update changelog * Update changelog * Resolve minor PR comments * Remove comments * Docs - Bump tqdm from 4.65.0 to 4.66.3 in /docs/sphinx (ROCm#1339) Bumps [tqdm](https://github.com/tqdm/tqdm) from 4.65.0 to 4.66.3. - [Release notes](https://github.com/tqdm/tqdm/releases) - [Commits](tqdm/tqdm@v4.65.0...v4.66.3) --- updated-dependencies: - dependency-name: tqdm dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * Docs - Bump jinja2 from 3.1.3 to 3.1.4 in /docs/sphinx (ROCm#1340) Bumps [jinja2](https://github.com/pallets/jinja) from 3.1.3 to 3.1.4. - [Release notes](https://github.com/pallets/jinja/releases) - [Changelog](https://github.com/pallets/jinja/blob/main/CHANGES.rst) - [Commits](pallets/jinja@3.1.3...3.1.4) --- updated-dependencies: - dependency-name: jinja2 dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * Find Half - Fix (ROCm#1341) * MIVisionX Setup - Updates (ROCm#1343) * SWDEV-459739 - Remove the package obsolete setting (ROCm#1345) The package was obsoleting itself and was causing upgrade issues. Removed the same. * Fix the layout issue with spec * Add layouts for Audio in vxTensorLayout Remove spectrogram layout param and pass layout in descriptor * Check the validity of pointers * Audio PR - Augmentation support [ Spectrogram ] (ROCm#1319) * Bump rocm-docs-core[api_reference] from 0.34.0 to 0.34.2 in /docs/sphinx (ROCm#1286) Bumps [rocm-docs-core[api_reference]](https://github.com/RadeonOpenCompute/rocm-docs-core) from 0.34.0 to 0.34.2. - [Release notes](https://github.com/RadeonOpenCompute/rocm-docs-core/releases) - [Changelog](https://github.com/ROCm/rocm-docs-core/blob/develop/CHANGELOG.md) - [Commits](ROCm/rocm-docs-core@v0.34.0...v0.34.2) --- updated-dependencies: - dependency-name: rocm-docs-core[api_reference] dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * Bump cryptography from 42.0.0 to 42.0.2 in /docs/sphinx (ROCm#1289) Bumps [cryptography](https://github.com/pyca/cryptography) from 42.0.0 to 42.0.2. - [Changelog](https://github.com/pyca/cryptography/blob/main/CHANGELOG.rst) - [Commits](pyca/cryptography@42.0.0...42.0.2) --- updated-dependencies: - dependency-name: cryptography dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * Add PreEmphasis filter support * Fix ROI - change from xy to width & height * Adding openvx changes for downmix node * Audio Augmentations 1 PR - NSR and Spectrogram * Revert "Adding openvx changes for downmix node" This reverts commit d53f81d. * PR comments resolution in the PreEmphais Filter * Minor Changes * Change the borderType enum to int32 from uint32 dtype * Fix validation of preemphasis * Remove the memcopy of the src and dest rois as it can be handled in the rocAL - since the src and dst rois remain same * Formatting change - minor * Remove NSR * Minor formatting changes * Minor fix * Minor update - remove the 2nd instance of preemphasis filter * Enum dtype - change from uint to int * Remove roi_tensor_ptr_dst as its unused after latest changes * Remove the dst_roi arg from vxExtRppPreemphasisFilter call as its unused * Add MFB to MIVisisonX * Revert "Add MFB to MIVisisonX" This reverts commit dc4200b. * Resolve the PR comments * Change the dims[0] and dims[1] positioning for Spectrogram and AudioFillDescPointers * Change function name to camelCase * Revert "Change the dims[0] and dims[1] positioning for Spectrogram and AudioFillDescPointers" This reverts commit 886d6af. * Fix Spectrogram * Docs - update TOC for API Ref (ROCm#1327) * Bump rocm-docs-core[api_reference] from 0.38.0 to 0.38.1 in /docs/sphinx (ROCm#1328) Bumps [rocm-docs-core[api_reference]](https://github.com/RadeonOpenCompute/rocm-docs-core) from 0.38.0 to 0.38.1. - [Release notes](https://github.com/RadeonOpenCompute/rocm-docs-core/releases) - [Changelog](https://github.com/ROCm/rocm-docs-core/blob/develop/CHANGELOG.md) - [Commits](ROCm/rocm-docs-core@v0.38.0...v0.38.1) --- updated-dependencies: - dependency-name: rocm-docs-core[api_reference] dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * Update CHANGELOG.md * Update CHANGELOG.md * Documents - Bump idna from 3.4 to 3.7 in /docs/sphinx (ROCm#1330) Bumps [idna](https://github.com/kjd/idna) from 3.4 to 3.7. - [Release notes](https://github.com/kjd/idna/releases) - [Changelog](https://github.com/kjd/idna/blob/master/HISTORY.rst) - [Commits](kjd/idna@v3.4...v3.7) --- updated-dependencies: - dependency-name: idna dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * Update changelog * Resolve minor PR comments * Remove comments * Fix the layout issue with spec * Check the validity of pointers --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: root <root@ixt-sjc2-54.local.lan> Co-authored-by: Swetha B S <swetha@mutlicorewareinc.com> Co-authored-by: SundarRajan28 <sundarrajan@multicorewareinc.com> Co-authored-by: Swetha B S <swetha@multiocrewareinc.com> Co-authored-by: randyh62 <42045079+randyh62@users.noreply.github.com> * Introduce API to obtain RPP layout * Add comments * Resolving review comments * Resolving review comments * Minor changes * Removing if blocks for CPU device checks * Use RPP_AUDIO flag to disable RPP audio calls * Add RPP_AUDIO flag for RPP audio API * Add Audio flag for PreEmphasis filter * Rename map variable --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: root <root@ixt-sjc2-54.local.lan> Co-authored-by: Swetha B S <swetha@mutlicorewareinc.com> Co-authored-by: Swetha B S <swetha@multicorewareinc.com> Co-authored-by: SundarRajan28 <sundarrajan@multicorewareinc.com> Co-authored-by: Swetha B S <swetha@multiocrewareinc.com> Co-authored-by: randyh62 <42045079+randyh62@users.noreply.github.com> Co-authored-by: swetha097 <59434434+swetha097@users.noreply.github.com> Co-authored-by: Kiriti Gowda <kiritigowda@gmail.com> Co-authored-by: raramakr <91213141+raramakr@users.noreply.github.com> Co-authored-by: fiona-gladwin <fionagladwin@multicorewareinc.com> Co-authored-by: Sundar Rajan Vaithiyanathan <99159823+SundarRajan28@users.noreply.github.com> Co-authored-by: Lakshmi Kumar <lakshmi.kumar@amd.com>

SundarRajan28 added 3 commits March 15, 2024 17:16

Adding openvx extensions for to_decibels and normalize augmentations

e63809c

Commenting out normalize kernel

589c823

Adding openvx extensions for resample and operator overloading nodes

7f77d64

SundarRajan28 requested review from fiona-gladwin and sampath1117 March 18, 2024 15:08

SundarRajan28 self-assigned this Mar 18, 2024

SundarRajan28 added 2 commits March 26, 2024 12:59

Merge branch 'swbs_m3/audio/pr6' into swbs_m5/audio/pr8

7c47fdd

Removing normalize node and adding formatting changes

6919be6

SundarRajan28 changed the base branch from swbs_m4/audio/pr7 to swbs_m3/audio/pr6 March 26, 2024 13:36

sampath1117 reviewed Mar 26, 2024

View reviewed changes

Merge branch 'swbs_m3/audio/pr6' into swbs_m5/audio/pr8

0a4f5b2

fiona-gladwin reviewed Mar 27, 2024

View reviewed changes

Adding resampling window calculation in openvx

ff041d4

SundarRajan28 changed the title ~~Audio PR 9 - Resample + Operator Overloading Nodes~~ Audio PR 8 - Resample + Operator Overloading Nodes Mar 29, 2024

Updating resample node with latest changes

32304a1

swetha097 added the audio pr label Mar 29, 2024

Resolving review comments

2977423

sampath1117 reviewed Apr 1, 2024

View reviewed changes

Fixing ROI issues with TensorTensorAdd node

6844edf

fiona-gladwin requested changes Apr 1, 2024

View reviewed changes

SundarRajan28 added 2 commits April 1, 2024 18:04

Resolving review comments

80568eb

Modifying pSrcROI in Resample openvx

02e9f66

SundarRajan28 force-pushed the swbs_m5/audio/pr8 branch from f2afb28 to 02e9f66 Compare April 2, 2024 05:35

SundarRajan28 and others added 7 commits April 2, 2024 14:03

Adding comments for tensor tensor add openvx node

0fe6488

Merge branch 'swbs_m3/audio/pr6' into swbs_m5/audio/pr8

cbcac89

Update changelog

e1d7f8e

Merge branch 'swbs_m3/audio/pr6' into swbs_m5/audio/pr8

b54cd21

Merge branch 'swbs_m3/audio/pr6' into swbs_m5/audio/pr8

e54a877

Merge branch 'swbs_m3/audio/pr6' into swbs_m5/audio/pr8

954eaa9

Merge branch 'swbs_m3/audio/pr6' into swbs_m5/audio/pr8

99c7303

SundarRajan28 added 2 commits May 13, 2024 08:34

Merge branch 'swbs_m3/audio/pr6' into swbs_m5/audio/pr8

b95c25a

Resolving review comments

2090e0b

sampath1117 reviewed May 14, 2024

View reviewed changes

SundarRajan28 and others added 13 commits May 15, 2024 04:06

Resolving review comments

1d1baeb

Merge branch 'swbs_m3/audio/pr6' into swbs_m5/audio/pr8

4b8ac7f

Merge branch 'swbs_m3/audio/pr6' into swbs_m5/audio/pr8

2502bc8

Merge branch 'swbs_m3/audio/pr6' into swbs_m5/audio/pr8

8bdad24

Merge branch 'swbs_m3/audio/pr6' into swbs_m5/audio/pr8

cc9c1ce

Add RPP_AUDIO flag for RPP audio API

1d0382c

Merge branch 'swbs_m3/audio/pr6' into swbs_m5/audio/pr8

2613545

Merge branch 'swbs_m3/audio/pr6' into swbs_m5/audio/pr8

fe22146

Merge branch 'develop' into swbs_m5/audio/pr8

26360e9

Merge branch 'swbs_m3/audio/pr6' into swbs_m5/audio/pr8

9832085

Merge branch 'swbs_m3/audio/pr6' into swbs_m5/audio/pr8

5bb4fd3

Merge remote-tracking branch 'upstream/develop' into swbs_m5/audio/pr8

2540b9e

Audio PR 8 - Resample + Operator Overloading Nodes #34

Are you sure you want to change the base?

Audio PR 8 - Resample + Operator Overloading Nodes #34

Conversation

SundarRajan28 commented Mar 18, 2024

sampath1117 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

sampath1117 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

fiona-gladwin left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

sampath1117 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment