#include "Annotations/PadData.h"
#include "Annotations/ShapeInferenceData.h"
#include "Annotations/StrideData.h"
+#include "Annotations/WindowData.h"
#include "Dialect/TFNodes.h"
#include <loco.h>
bool fix_padding(moco::tf::TFAvgPool *node)
{
- // Nothing to do with padding
- return false;
+ LOGGER(l);
+
+ auto pad_data_c = node->annot<PadData>();
+ if (pad_data_c != nullptr)
+ {
+ // padding conversion is already done
+ return false;
+ }
+
+ auto ofm_shapedata = node->annot<ShapeInferenceData>();
+ if (ofm_shapedata == nullptr)
+ {
+ // need output shape to calculate padding values
+ return false;
+ }
+ auto value = node->value();
+ assert(value != nullptr);
+ auto value_shapedata = value->annot<ShapeInferenceData>();
+ if (value_shapedata == nullptr)
+ {
+ // need input shape to calculate padding values
+ return false;
+ }
+ auto stride_data = node->annot<StrideData>();
+ if (stride_data == nullptr)
+ {
+ // need stride_data from FixShape
+ return false;
+ }
+ auto window_data = node->annot<WindowData>();
+ if (window_data == nullptr)
+ {
+ // need window_data from FixShape
+ return false;
+ }
+
+ auto padding = node->padding();
+ assert(padding == "VALID" || padding == "SAME");
+ assert(ofm_shapedata->rank() == 4);
+ assert(value_shapedata->rank() == 4);
+
+ auto value_feature_shape = as_feature_shape(*value_shapedata, node->data_layout());
+ auto ofm_feature_shape = as_feature_shape(*ofm_shapedata, node->data_layout());
+
+ uint32_t input_height = value_feature_shape.height().value();
+ uint32_t input_width = value_feature_shape.width().value();
+ uint32_t stride_height = stride_data->stride()->vertical();
+ uint32_t stride_width = stride_data->stride()->horizontal();
+ uint32_t window_height = window_data->window()->vertical();
+ uint32_t window_width = window_data->window()->horizontal();
+ uint32_t output_height = ofm_feature_shape.height().value();
+ uint32_t output_width = ofm_feature_shape.width().value();
+ uint32_t dilation_height = 1; // dilation for AvgPool is 1
+ uint32_t dilation_width = 1;
+ uint32_t effective_window_height = dilation_height * (window_height - 1) + 1;
+ uint32_t effective_window_width = dilation_width * (window_width - 1) + 1;
+ // calculate padding height, width
+ int32_t i_height = (output_height - 1) * stride_height + effective_window_height - input_height;
+ int32_t i_width = (output_width - 1) * stride_width + effective_window_width - input_width;
+ uint32_t height = i_height >= 0 ? i_height : 0U;
+ uint32_t width = i_width >= 0 ? i_width : 0U;
+
+ // annotation of pad data
+ auto pad_data = stdex::make_unique<PadData>();
+
+ pad_data->pad()->top(height / 2);
+ pad_data->pad()->bottom(height - pad_data->pad()->top());
+ pad_data->pad()->left(width / 2);
+ pad_data->pad()->right(width - pad_data->pad()->left());
+
+ node->annot(std::move(pad_data));
+
+ {
+ auto pad_data = node->annot<PadData>();
+ assert(pad_data != nullptr);
+
+ // clang-format off
+ INFO(l) << "Fix TFAvgPool pad "
+ << "= T " << pad_data->pad()->top()
+ << ", L " << pad_data->pad()->left()
+ << ", B " << pad_data->pad()->bottom()
+ << ", R " << pad_data->pad()->right() << std::endl;
+ // clang-format on
+ }
+ return true;
}
bool fix_padding(moco::tf::TFBiasAdd *node)
#include "Annotations/PaddingData.h"
#include "Annotations/ShapeInferenceData.h"
#include "Annotations/StrideData.h"
+#include "Annotations/WindowData.h"
#include "Dialect/TFNodes.h"
#include <loco.h>
bool fix_shape(moco::tf::TFAvgPool *node)
{
- // TODO implement
- throw std::runtime_error("NYI fix_shape TFAvgPool");
+ LOGGER(l);
+
+ auto shapedata = node->annot<ShapeInferenceData>();
+ if (shapedata != nullptr)
+ {
+ // shape inference is already done for TFAvgPool
+ return false;
+ }
+ auto value = node->value();
+ auto value_shapedata = value->annot<ShapeInferenceData>();
+ if (value_shapedata == nullptr)
+ {
+ // input node shape inference is not ready
+ return false;
+ }
+
+ auto padding = node->padding();
+ assert(padding == "VALID" || padding == "SAME");
+
+ // TODO move this to some new Transformation...
+ {
+ {
+ auto stride_data = node->annot<StrideData>();
+ assert(stride_data == nullptr);
+ }
+ auto stride_data = stdex::make_unique<StrideData>();
+ auto strides = node->strides();
+ auto data_layout = plier::tf::as_data_layout(node->data_layout());
+ if (data_layout == plier::tf::DataLayout::NHWC)
+ {
+ stride_data->stride()->vertical(strides[1]);
+ stride_data->stride()->horizontal(strides[2]);
+ }
+ else if (data_layout == plier::tf::DataLayout::NCHW)
+ {
+ stride_data->stride()->vertical(strides[2]);
+ stride_data->stride()->horizontal(strides[3]);
+ }
+ node->annot(std::move(stride_data));
+
+ {
+ auto window_data = node->annot<WindowData>();
+ assert(window_data == nullptr);
+ }
+ auto window_data = stdex::make_unique<WindowData>();
+ auto ksize = node->ksize();
+ if (data_layout == plier::tf::DataLayout::NHWC)
+ {
+ window_data->window()->vertical(ksize[1]);
+ window_data->window()->horizontal(ksize[2]);
+ }
+ else if (data_layout == plier::tf::DataLayout::NCHW)
+ {
+ window_data->window()->vertical(ksize[2]);
+ window_data->window()->horizontal(ksize[3]);
+ }
+ node->annot(std::move(window_data));
+ }
+
+ auto value_feature_shape = as_feature_shape(*value_shapedata, node->data_layout());
+
+ auto stride_data = node->annot<StrideData>();
+ assert(stride_data != nullptr);
+ auto window_data = node->annot<WindowData>();
+ assert(window_data != nullptr);
+
+ uint32_t input_height = value_feature_shape.height().value();
+ uint32_t input_width = value_feature_shape.width().value();
+ uint32_t stride_height = stride_data->stride()->vertical();
+ uint32_t stride_width = stride_data->stride()->horizontal();
+ uint32_t window_height = window_data->window()->vertical();
+ uint32_t window_width = window_data->window()->horizontal();
+ uint32_t dilation_height = 1; // dilation is 1
+ uint32_t dilation_width = 1;
+ uint32_t effective_window_height = dilation_height * (window_height - 1) + 1;
+ uint32_t effective_window_width = dilation_width * (window_width - 1) + 1;
+ uint32_t output_height;
+ uint32_t output_width;
+
+ if (padding == "VALID")
+ {
+ output_height = (input_height + stride_height - effective_window_height) / stride_height;
+ output_width = (input_width + stride_width - effective_window_width) / stride_width;
+ }
+ else if (padding == "SAME")
+ {
+ output_height = (input_height + stride_height - 1) / stride_height;
+ output_width = (input_width + stride_width - 1) / stride_width;
+ }
+
+ loco::FeatureShape ofm_feature_shape;
+ ofm_feature_shape.count() = value_feature_shape.count();
+ ofm_feature_shape.height() = output_height;
+ ofm_feature_shape.width() = output_width;
+ ofm_feature_shape.depth() = value_feature_shape.depth();
+
+ auto shape_data = stdex::make_unique<ShapeInferenceData>();
+ as_tensor_shape(*shape_data.get(), ofm_feature_shape, node->data_layout());
+ node->annot(std::move(shape_data));
+
+ INFO(l) << "Fix TFAvgPool shape = ifm" << value_feature_shape << " --> ofm" << ofm_feature_shape;
+
+ return true;
}
bool fix_shape(moco::tf::TFBiasAdd *node)