Method Forward

src/model/layer/convolution.cc:107–136

\copydoc Layer::Forward(int flag, const Tensor&)

/// \copydoc Layer::Forward(int flag, const Tensor&)
const Tensor Convolution::Forward(int flag, const Tensor &input) {
  CHECK(buf_.empty());
  CHECK_EQ(input.device()->lang(), kCpp);
  CHECK_EQ(input.nDim(), 4u);
  if (flag & kTrain) buf_.push(input);
  size_t batchsize = input.shape(0);
  size_t imagesize = input.Size() / batchsize;
  // TODO(wangwei) update the layer config if the input sample shape changes
  CHECK(input.shape(1) == channels_ && input.shape(2) == height_ &&
      input.shape(3) == width_) << "input sample shape should not change";
  DataType dtype = input.data_type();
  auto dev = input.device();
  Shape shape{batchsize, num_filters_, conv_height_, conv_width_};
  Tensor output(shape, dev, dtype);
  Tensor col_data(Shape{col_height_, col_width_});
  float *data_col = new float[col_height_ * col_width_];
  auto in_data = input.data<float>();
  for (size_t b = 0; b < batchsize; b++) {
    Im2col(in_data + b * imagesize, channels_, height_, width_, kernel_h_,
        kernel_w_, pad_h_, pad_w_, stride_h_, stride_w_, data_col);
    col_data.CopyDataFromHostPtr(data_col, col_height_ * col_width_);
    Tensor each = Mult(weight_, col_data);
    if (bias_term_) {
      AddColumn(bias_, &each);
    }
    CopyDataToFrom(&output, each, each.Size(), b * each.Size());
  }
  delete[] data_col;
  return output;
}

/// \copydoc Layer::Backward(int, const Tensor&, const Tensor&);
const std::pair<Tensor, vector<Tensor>> Convolution::Backward(