385 bool reshapePrepare(const Shape& input, const int32_t* targetDims, const int32_t targetDimsSize,  in reshapePrepare()  argument
391     int32_t numInputElements = (int32_t)getNumberOfElements(input);  in reshapePrepare()
414     output->type = input.type;  in reshapePrepare()
416     output->offset = input.offset;  in reshapePrepare()
417     output->scale = input.scale;  in reshapePrepare()
422 bool depthToSpacePrepare(const Shape& input, int32_t blockSize, Shape* output) {  in depthToSpacePrepare()  argument
423     NN_OPS_CHECK(getNumberOfDimensions(input) == 4);  in depthToSpacePrepare()
426     uint32_t batches = getSizeOfDimension(input, 0);  in depthToSpacePrepare()
427     uint32_t height = getSizeOfDimension(input, 1);  in depthToSpacePrepare()
428     uint32_t width = getSizeOfDimension(input, 2);  in depthToSpacePrepare()
429     uint32_t channels = getSizeOfDimension(input, 3);  in depthToSpacePrepare()
432     output->type = input.type;  in depthToSpacePrepare()
435     output->offset = input.offset;  in depthToSpacePrepare()
436     output->scale = input.scale;  in depthToSpacePrepare()
441 bool spaceToDepthPrepare(const Shape& input, int32_t blockSize, Shape* output) {  in spaceToDepthPrepare()  argument
442     NN_OPS_CHECK(getNumberOfDimensions(input) == 4);  in spaceToDepthPrepare()
445     uint32_t batches = getSizeOfDimension(input, 0);  in spaceToDepthPrepare()
446     uint32_t height = getSizeOfDimension(input, 1);  in spaceToDepthPrepare()
447     uint32_t width = getSizeOfDimension(input, 2);  in spaceToDepthPrepare()
448     uint32_t channels = getSizeOfDimension(input, 3);  in spaceToDepthPrepare()
453     output->type = input.type;  in spaceToDepthPrepare()
456     output->offset = input.offset;  in spaceToDepthPrepare()
457     output->scale = input.scale;  in spaceToDepthPrepare()
507 bool padPrepare(const Shape& input, const int32_t* paddingsData, const Shape& paddingsShape,  in padPrepare()  argument
509     uint32_t numInputDims = getNumberOfDimensions(input);  in padPrepare()
523         outDims[i] = beforePadding + getSizeOfDimension(input, i) + afterPadding;  in padPrepare()
525     output->type = input.type;  in padPrepare()
527     output->offset = input.offset;  in padPrepare()
528     output->scale = input.scale;  in padPrepare()
533 bool batchToSpacePrepare(const Shape& input, const int32_t* blockSizeData,  in batchToSpacePrepare()  argument
536     NN_OPS_CHECK(getNumberOfDimensions(input) == 4);  in batchToSpacePrepare()
544     uint32_t batches = getSizeOfDimension(input, 0);  in batchToSpacePrepare()
545     uint32_t height = getSizeOfDimension(input, 1);  in batchToSpacePrepare()
546     uint32_t width = getSizeOfDimension(input, 2);  in batchToSpacePrepare()
547     uint32_t channels = getSizeOfDimension(input, 3);  in batchToSpacePrepare()
550     output->type = input.type;  in batchToSpacePrepare()
553     output->offset = input.offset;  in batchToSpacePrepare()
554     output->scale = input.scale;  in batchToSpacePrepare()
559 bool spaceToBatchPrepare(const Shape& input, const int32_t* blockSizeData,  in spaceToBatchPrepare()  argument
563     NN_OPS_CHECK(getNumberOfDimensions(input) == 4);  in spaceToBatchPrepare()
577     uint32_t batches = getSizeOfDimension(input, 0);  in spaceToBatchPrepare()
578     uint32_t height = getSizeOfDimension(input, 1);  in spaceToBatchPrepare()
579     uint32_t width = getSizeOfDimension(input, 2);  in spaceToBatchPrepare()
580     uint32_t channels = getSizeOfDimension(input, 3);  in spaceToBatchPrepare()
588     output->type = input.type;  in spaceToBatchPrepare()
592     output->offset = input.offset;  in spaceToBatchPrepare()
593     output->scale = input.scale;  in spaceToBatchPrepare()
598 bool meanPrepare(const Shape& input, const int32_t* axisData, const Shape& axisShape, bool keepDims,  in meanPrepare()  argument
604     int32_t numInputDims = static_cast<int32_t>(getNumberOfDimensions(input));  in meanPrepare()
621                 outDims[idx] = getSizeOfDimension(input, idx);  in meanPrepare()
658                 outDims[idx - numSkipAxis] = getSizeOfDimension(input, idx);  in meanPrepare()
668     output->type = input.type;  in meanPrepare()
669     output->offset = input.offset;  in meanPrepare()
670     output->scale = input.scale;  in meanPrepare()
675 bool argMinMaxPrepare(const Shape& input, int32_t axis, Shape* output) {  in argMinMaxPrepare()  argument
676     NN_CHECK(handleNegativeAxis(input, &axis));  in argMinMaxPrepare()
682     if (getNumberOfDimensions(input) > 1) {  in argMinMaxPrepare()
683         output->dimensions.reserve(getNumberOfDimensions(input) - 1);  in argMinMaxPrepare()
684         output->dimensions.insert(output->dimensions.end(), input.dimensions.begin(),  in argMinMaxPrepare()
685                                   input.dimensions.begin() + axis);  in argMinMaxPrepare()
686         output->dimensions.insert(output->dimensions.end(), input.dimensions.begin() + axis + 1,  in argMinMaxPrepare()
687                                   input.dimensions.end());  in argMinMaxPrepare()
695 bool splitPrepare(const Shape& input, int32_t axis, int32_t numOutputs,  in splitPrepare()  argument
697     NN_CHECK(handleNegativeAxis(input, &axis));  in splitPrepare()
699     const int32_t sizeOfAxisToSplit = input.dimensions[axis];  in splitPrepare()
704         output->at(i).type = input.type;  in splitPrepare()
705         output->at(i).dimensions = input.dimensions;  in splitPrepare()
707         output->at(i).offset = input.offset;  in splitPrepare()
708         output->at(i).scale = input.scale;  in splitPrepare()
713 bool groupedConvPrepare(const Shape& input, const Shape& filter, const Shape& bias,  in groupedConvPrepare()  argument
718         NN_OPS_CHECK(input.type == OperandType::TENSOR_QUANT8_ASYMM ||  in groupedConvPrepare()
719                      input.type == OperandType::TENSOR_QUANT8_ASYMM_SIGNED);  in groupedConvPrepare()
721         NN_OPS_CHECK(input.type == filter.type);  in groupedConvPrepare()
723     if (input.type == OperandType::TENSOR_QUANT8_ASYMM ||  in groupedConvPrepare()
724         input.type == OperandType::TENSOR_QUANT8_ASYMM_SIGNED) {  in groupedConvPrepare()
727         NN_OPS_CHECK(input.type == bias.type);  in groupedConvPrepare()
729     NN_OPS_CHECK(getNumberOfDimensions(input) == 4);  in groupedConvPrepare()
735     NN_OPS_CHECK(getSizeOfDimension(filter, 3) * numGroups == getSizeOfDimension(input, 3));  in groupedConvPrepare()
739     uint32_t width = getSizeOfDimension(input, 2);  in groupedConvPrepare()
740     uint32_t height = getSizeOfDimension(input, 1);  in groupedConvPrepare()
743     uint32_t batches = getSizeOfDimension(input, 0);  in groupedConvPrepare()
755     output->type = input.type;  in groupedConvPrepare()