diff options
Diffstat (limited to 'compute/cker/include/cker/operation/TransposeConv.h')
| -rw-r--r-- | compute/cker/include/cker/operation/TransposeConv.h | 135 |
1 files changed, 135 insertions, 0 deletions
diff --git a/compute/cker/include/cker/operation/TransposeConv.h b/compute/cker/include/cker/operation/TransposeConv.h new file mode 100644 index 000000000..535fe86cf --- /dev/null +++ b/compute/cker/include/cker/operation/TransposeConv.h @@ -0,0 +1,135 @@ +/* + * Copyright (c) 2019 Samsung Electronics Co., Ltd. All Rights Reserved + * Copyright 2017 The TensorFlow Authors. All Rights Reserved. + * + * Licensed under the Apache License, Version 2.0 (the "License"); + * you may not use this file except in compliance with the License. + * You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +#ifndef __NNFW_CKER_TRANSPOSE_CONV_H__ +#define __NNFW_CKER_TRANSPOSE_CONV_H__ + +#include "cker/Shape.h" +#include "cker/Types.h" +#include "cker/Utils.h" + +namespace nnfw +{ +namespace cker +{ + +struct TransposeConvParams +{ + PaddingType padding_type; + PaddingValues padding_values; + // TODO(starka): This was just "stride", so check that width+height is OK. + int16_t stride_width; + int16_t stride_height; + int16_t dilation_width_factor; + int16_t dilation_height_factor; + // uint8_t inference params. + // TODO(b/65838351): Use smaller types if appropriate. + int32_t input_offset; + int32_t weights_offset; + int32_t output_offset; + int32_t output_multiplier; + int output_shift; + // uint8_t, etc, activation params. + int32_t quantized_activation_min; + int32_t quantized_activation_max; + // float activation params. + float float_activation_min; + float float_activation_max; +}; + +inline void TransposeConv(const TransposeConvParams ¶ms, const Shape &input_shape, + const float *input_data, const Shape &filter_shape, + const float *filter_data, const Shape &output_shape, float *output_data) +{ + + const int stride_width = params.stride_width; + const int stride_height = params.stride_height; + const int pad_width = params.padding_values.width; + const int pad_height = params.padding_values.height; + + assert(input_shape.DimensionsCount() == 4); + assert(filter_shape.DimensionsCount() == 4); + assert(output_shape.DimensionsCount() == 4); + + const int batches = MatchingDim(input_shape, 0, output_shape, 0); + const int input_depth = MatchingDim(input_shape, 3, filter_shape, 3); + const int output_depth = MatchingDim(filter_shape, 0, output_shape, 3); + const int input_height = input_shape.Dims(1); + const int input_width = input_shape.Dims(2); + const int filter_height = filter_shape.Dims(1); + const int filter_width = filter_shape.Dims(2); + const int output_height = output_shape.Dims(1); + const int output_width = output_shape.Dims(2); + + // Although transpose convolution simplifies to convolution with transposed + // weights for strides of 1, non-unitary striding complicates matters. To + // keep this reference implementation as clear as possible, we use a + // "scatter" access pattern, where we loop through all the input elements, + // computing their influence on the output, rather than looping through the + // output elements in the typical "gather" access pattern of a conv. We + // therefore must initialize the output array to zero. + const int num_elements = output_shape.FlatSize(); + for (int i = 0; i < num_elements; i++) + { + output_data[i] = 0.0f; + } + + // Loop through input elements one at a time. + for (int batch = 0; batch < batches; ++batch) + { + for (int in_y = 0; in_y < input_height; ++in_y) + { + for (int in_x = 0; in_x < input_width; ++in_x) + { + for (int in_channel = 0; in_channel < input_depth; ++in_channel) + { + // Loop through the output elements it will influence + const int out_x_origin = (in_x * stride_width) - pad_width; + const int out_y_origin = (in_y * stride_height) - pad_height; + for (int filter_y = 0; filter_y < filter_height; ++filter_y) + { + for (int filter_x = 0; filter_x < filter_width; ++filter_x) + { + for (int out_channel = 0; out_channel < output_depth; ++out_channel) + { + // Compute output element location + const int out_x = out_x_origin + filter_x; + const int out_y = out_y_origin + filter_y; + // We cannot accumulate out of bounds + if ((out_x >= 0) && (out_x < output_width) && (out_y >= 0) && + (out_y < output_height)) + { + float input_value = + input_data[Offset(input_shape, batch, in_y, in_x, in_channel)]; + float filter_value = filter_data[Offset(filter_shape, out_channel, filter_y, + filter_x, in_channel)]; + output_data[Offset(output_shape, batch, out_y, out_x, out_channel)] += + input_value * filter_value; + } + } + } + } + } + } + } + } +} + +} // namespace cker +} // namespace nnfw + +#endif // __NNFW_CKER_TRANSPOSE_CONV_H__ |