Overview

Implement convolution operator in CNN step by step for ease of understanding, regardless of efficiency.

Currently including:

conv2d_with_stride
conv2d_with_stride_padding
conv2d_with_stride_padding_dilation
conv2d_with_stride_padding_dilation_groups
- mode='plain': plain implementation.
- mode='im2col': use im2col borrowed from Caffe and matmul implemented by ourselves.
- mode='im2col_v2': use im2col borrowed from Caffe and np.dot.

We also compare the results of our implementions with Pytorch's.

Dependancy

mkdir -p build
cmake ..
make
cd ..
python test.py

Name		Name	Last commit message	Last commit date
Latest commit History 7 Commits
.gitignore		.gitignore
CMakeLists.txt		CMakeLists.txt
README.md		README.md
conv_impl.cpp		conv_impl.cpp
diy_conv.py		diy_conv.py
test.py		test.py