arm_compute v17.06
diff --git a/src/runtime/NEON/functions/NEConvolutionLayer.cpp b/src/runtime/NEON/functions/NEConvolutionLayer.cpp
index aae4a67..bd688cf 100644
--- a/src/runtime/NEON/functions/NEConvolutionLayer.cpp
+++ b/src/runtime/NEON/functions/NEConvolutionLayer.cpp
@@ -33,33 +33,93 @@
using namespace arm_compute;
-NEConvolutionLayer::NEConvolutionLayer()
- : _input_im2col_kernel(), _input_interleave_kernel(), _weights_reshape_kernel(), _weights_transposed_kernel(), _mm_kernel(), _output_col2im_kernel(), _input_im2col_reshaped(),
- _input_interleaved_reshaped(), _weights_reshaped(), _weights_transposed(), _gemm_output(), _is_first_run(false), _has_bias(false)
+NEConvolutionLayerReshapeWeights::NEConvolutionLayerReshapeWeights()
+ : _weights_reshape_kernel(), _weights_transposed_kernel(), _weights_reshaped(), _transpose1xW(false)
{
}
-void NEConvolutionLayer::configure(const ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, const PadStrideInfo &conv_info)
+void NEConvolutionLayerReshapeWeights::configure(const ITensor *weights, const ITensor *biases, ITensor *output, bool transpose1xW)
{
- ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(input, 1, DataType::F32);
- ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(weights, 1, DataType::F32);
- ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(output, 1, DataType::F32);
- ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_TYPES(input, weights, output);
- ARM_COMPUTE_ERROR_ON(weights->info()->dimension(2) != input->info()->dimension(2));
+ ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(weights, 1, DataType::QS8, DataType::F32);
+ ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(output, 1, DataType::QS8, DataType::F32);
+ ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_TYPES(weights, output);
+ ARM_COMPUTE_ERROR_ON_MISMATCHING_FIXED_POINT(weights, output);
ARM_COMPUTE_ERROR_ON(weights->info()->num_dimensions() > 4);
if(biases != nullptr)
{
- ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(biases, 1, DataType::F32);
- ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_TYPES(input, biases);
+ ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(biases, 1, DataType::QS8, DataType::F32);
+ ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_TYPES(weights, biases);
+ ARM_COMPUTE_ERROR_ON_MISMATCHING_FIXED_POINT(weights, biases);
ARM_COMPUTE_ERROR_ON(biases->info()->dimension(0) != weights->info()->dimension(3));
ARM_COMPUTE_ERROR_ON(biases->info()->num_dimensions() > 1);
}
- _has_bias = (biases != nullptr);
- _is_first_run = true;
+ // Check if bias are present, if yes they will be embedded to the weights matrix
+ const bool _has_bias = (biases != nullptr);
- // Get parameters for conv_info
+ _transpose1xW = transpose1xW;
+
+ if(transpose1xW)
+ {
+ // Create tensor to store the reshaped weights
+ const unsigned int mat_weights_cols = weights->info()->dimension(3);
+ const unsigned int mat_weights_rows = weights->info()->dimension(0) * weights->info()->dimension(1) * weights->info()->dimension(2) + (_has_bias ? 1 : 0);
+ TensorShape shape_wr(mat_weights_cols, mat_weights_rows);
+ TensorInfo info_wr(shape_wr, 1, weights->info()->data_type(), weights->info()->fixed_point_position());
+
+ _weights_reshaped.allocator()->init(info_wr);
+ _weights_reshape_kernel.configure(weights, biases, &_weights_reshaped);
+ _weights_transposed_kernel.configure(&_weights_reshaped, output);
+ _weights_reshaped.allocator()->allocate();
+ }
+ else
+ {
+ _weights_reshape_kernel.configure(weights, biases, output);
+ }
+}
+
+void NEConvolutionLayerReshapeWeights::run()
+{
+ NEScheduler::get().schedule(&_weights_reshape_kernel, 3);
+ if(_transpose1xW)
+ {
+ NEScheduler::get().schedule(&_weights_transposed_kernel, Window::DimY);
+ }
+}
+
+NEConvolutionLayer::NEConvolutionLayer()
+ : _input_im2col_kernel(), _input_interleave_kernel(), _reshape_weights(), _mm_kernel(), _output_col2im_kernel(), _input_im2col_reshaped(), _input_interleaved_reshaped(), _weights_reshaped(),
+ _gemm_output(), _has_bias(false), _is_fully_connected_convolution(false), _are_weights_reshaped(false)
+{
+}
+
+void NEConvolutionLayer::configure(const ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, const PadStrideInfo &conv_info, const WeightsInfo &weights_info)
+{
+ ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(input, 1, DataType::QS8, DataType::F32);
+ ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(weights, 1, DataType::QS8, DataType::F32);
+ ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(output, 1, DataType::QS8, DataType::F32);
+ ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_TYPES(input, weights, output);
+ ARM_COMPUTE_ERROR_ON_MISMATCHING_FIXED_POINT(input, weights, output);
+ ARM_COMPUTE_ERROR_ON(!weights_info.are_reshaped() && weights->info()->dimension(2) != input->info()->dimension(2));
+ ARM_COMPUTE_ERROR_ON(weights->info()->num_dimensions() > 4);
+
+ if(biases != nullptr)
+ {
+ ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(biases, 1, DataType::QS8, DataType::F32);
+ ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_TYPES(input, biases);
+ ARM_COMPUTE_ERROR_ON_MISMATCHING_FIXED_POINT(input, biases);
+ ARM_COMPUTE_ERROR_ON(!weights_info.are_reshaped() && biases->info()->dimension(0) != weights->info()->dimension(3));
+ ARM_COMPUTE_ERROR_ON(biases->info()->num_dimensions() > 1);
+ }
+
+ const DataType dt = input->info()->data_type();
+ const int fixed_point_position = input->info()->fixed_point_position();
+
+ _has_bias = (biases != nullptr);
+ _are_weights_reshaped = weights_info.are_reshaped();
+
+ // Get parameters from conv_info
unsigned int stride_x = 0;
unsigned int stride_y = 0;
unsigned int pad_x = 0;
@@ -70,21 +130,46 @@
// Get convolved dimensions
unsigned int conv_w = 0;
unsigned int conv_h = 0;
- std::tie(conv_w, conv_h) = scaled_dimensions(input->info()->dimension(0), input->info()->dimension(1), weights->info()->dimension(0),
+
+ const unsigned int kernel_width = (_are_weights_reshaped) ? weights_info.kernel_size() : weights->info()->dimension(0);
+ std::tie(conv_w, conv_h) = scaled_dimensions(input->info()->dimension(0), input->info()->dimension(1), kernel_width,
stride_x, stride_y, pad_x, pad_y, conv_info.round());
ARM_COMPUTE_ERROR_ON_MSG((output->info()->dimension(0) != conv_w) || (output->info()->dimension(1) != conv_h), "Output shape does not match the expected one");
- // Create tensor to store the reshaped weights
- const unsigned int mat_weights_cols = weights->info()->dimension(3);
- const unsigned int mat_weights_rows = weights->info()->dimension(0) * weights->info()->dimension(1) * weights->info()->dimension(2) + (_has_bias ? 1 : 0);
- TensorShape shape_wr(mat_weights_cols, mat_weights_rows);
- TensorInfo info_wr(shape_wr, 1, weights->info()->data_type());
- _weights_reshaped.allocator()->init(info_wr);
+ // Check if its a "fully connected" convolution
+ _is_fully_connected_convolution = ((conv_w == 1) && (conv_h == 1));
- // Create tensor to store transposed weights
- TensorShape shape_wt(mat_weights_rows * 4, static_cast<unsigned int>(std::ceil(mat_weights_cols / 4.f)));
- TensorInfo info_wt(shape_wt, 1, weights->info()->data_type());
- _weights_transposed.allocator()->init(info_wt);
+ unsigned int mat_weights_cols = weights->info()->dimension(3);
+ unsigned int mat_weights_rows = weights->info()->dimension(0) * weights->info()->dimension(1) * weights->info()->dimension(2) + (_has_bias ? 1 : 0);
+
+ // Reshape weights if needed
+ if(_are_weights_reshaped)
+ {
+ mat_weights_cols = output->info()->dimension(2);
+ const unsigned int quarter_reshaped_cols = weights->info()->dimension(0) / 4;
+ mat_weights_rows = (_has_bias ? 1 + quarter_reshaped_cols : quarter_reshaped_cols);
+ }
+ else
+ {
+ if(_is_fully_connected_convolution)
+ {
+ // Create tensor to store the reshaped weights
+ TensorShape shape_wr(mat_weights_cols, mat_weights_rows);
+ TensorInfo info_wr(shape_wr, 1, dt, fixed_point_position);
+ _weights_reshaped.allocator()->init(info_wr);
+ _reshape_weights.configure(weights, biases, &_weights_reshaped, false /* 1xW transpose */);
+ }
+ else
+ {
+ // Create tensor to store transposed weights
+ const float transpose_width = 16.0f / input->info()->element_size();
+ TensorShape shape_wt(mat_weights_rows * static_cast<unsigned int>(transpose_width), static_cast<unsigned int>(std::ceil(mat_weights_cols / transpose_width)));
+ TensorInfo info_wt(shape_wt, 1, dt, fixed_point_position);
+ _weights_reshaped.allocator()->init(info_wt);
+ _reshape_weights.configure(weights, biases, &_weights_reshaped, true /* 1xW transpose */);
+ }
+ weights = &_weights_reshaped;
+ }
// Create tensor to store im2col reshaped inputs
const unsigned int mat_input_cols = mat_weights_rows;
@@ -93,58 +178,69 @@
shape_im2col.set(0, mat_input_cols);
shape_im2col.set(1, mat_input_rows);
shape_im2col.set(2, 1);
- TensorInfo info_im2col(shape_im2col, 1, input->info()->data_type());
- _input_im2col_reshaped.allocator()->init(info_im2col);
+ _input_im2col_reshaped.allocator()->init(TensorInfo(shape_im2col, 1, dt, fixed_point_position));
- // Create tensor to prepare input tensor for GEMM
- TensorShape shape_interleaved = shape_im2col;
- shape_interleaved.set(0, shape_interleaved.x() * 4);
- shape_interleaved.set(1, std::ceil(shape_interleaved.y() / 4.f));
- TensorInfo info_interleaved(shape_interleaved, 1, input->info()->data_type());
- _input_interleaved_reshaped.allocator()->init(info_interleaved);
+ // Create tensor (interleave) to prepare input tensor for GEMM
+ if(!_is_fully_connected_convolution)
+ {
+ TensorShape shape_interleaved = shape_im2col;
+ shape_interleaved.set(0, shape_interleaved.x() * 4);
+ shape_interleaved.set(1, std::ceil(shape_interleaved.y() / 4.f));
+ _input_interleaved_reshaped.allocator()->init(TensorInfo(shape_interleaved, 1, dt, fixed_point_position));
+ }
// Create GEMM output tensor
TensorShape shape_gemm = _input_im2col_reshaped.info()->tensor_shape();
shape_gemm.set(0, mat_weights_cols);
shape_gemm.set(1, mat_input_rows);
- TensorInfo info_gemm(shape_gemm, 1, input->info()->data_type());
- _gemm_output.allocator()->init(info_gemm);
+ _gemm_output.allocator()->init(TensorInfo(shape_gemm, 1, dt, fixed_point_position));
// Configure kernels
_input_im2col_kernel.configure(input, &_input_im2col_reshaped, std::make_pair(conv_w, conv_h), conv_info, _has_bias);
- _input_interleave_kernel.configure(&_input_im2col_reshaped, &_input_interleaved_reshaped);
- _weights_reshape_kernel.configure(weights, biases, &_weights_reshaped);
- _weights_transposed_kernel.configure(&_weights_reshaped, &_weights_transposed);
- _mm_kernel.configure(&_input_interleaved_reshaped, &_weights_transposed, &_gemm_output, 1.0f);
+ if(_is_fully_connected_convolution)
+ {
+ _mm_kernel.configure(&_input_im2col_reshaped, weights, &_gemm_output, 1.0f);
+ }
+ else
+ {
+ _input_interleave_kernel.configure(&_input_im2col_reshaped, &_input_interleaved_reshaped);
+ _mm_kernel.configure(&_input_interleaved_reshaped, weights, &_gemm_output, 1.0f);
+ }
_output_col2im_kernel.configure(&_gemm_output, output, std::make_pair(conv_w, conv_h));
- // Allocate the tensors once the all configure methods have been called
- _weights_reshaped.allocator()->allocate();
- _weights_transposed.allocator()->allocate();
+ // Allocate intermediate tensor
+ if(!_are_weights_reshaped)
+ {
+ _weights_reshaped.allocator()->allocate();
+ }
_input_im2col_reshaped.allocator()->allocate();
- _input_interleaved_reshaped.allocator()->allocate();
+ if(!_is_fully_connected_convolution)
+ {
+ _input_interleaved_reshaped.allocator()->allocate();
+ }
_gemm_output.allocator()->allocate();
}
void NEConvolutionLayer::run()
{
// Run weights reshaping (Runs once for every configure)
- if(_is_first_run)
+ if(!_are_weights_reshaped)
{
- _is_first_run = false;
- NEScheduler::get().multithread(&_weights_reshape_kernel, 3);
- NEScheduler::get().multithread(&_weights_transposed_kernel);
+ _are_weights_reshaped = true;
+ _reshape_weights.run();
}
// Run input reshaping
- NEScheduler::get().multithread(&_input_im2col_kernel);
+ NEScheduler::get().schedule(&_input_im2col_kernel, Window::DimY);
+ if(!_is_fully_connected_convolution)
+ {
+ // Run interleave
+ NEScheduler::get().schedule(&_input_interleave_kernel, Window::DimY);
+ }
- // Run interleave
- NEScheduler::get().multithread(&_input_interleave_kernel);
-
- // Runs GEMM on reshaped matrices
- NEScheduler::get().multithread(&_mm_kernel);
+ // Runs matrix multiply on reshaped matrices
+ NEScheduler::get().schedule(&_mm_kernel, Window::DimY);
// Reshape output matrix
- NEScheduler::get().multithread(&_output_col2im_kernel);
+ NEScheduler::get().schedule(&_output_col2im_kernel, Window::DimY);
}