search

fengggli / gpu-computing-materials

A simple deep learning framework that optimizes task scheduling and memory usage on different CPU/GPU architectures.
1 stars 0 forks source link

Backward #25

Closed zkSNARK closed 5 years ago

zkSNARK commented 5 years ago

This commit makes a number of bugfixes, but is primarily focused on adding the backward function in the layer_conv.c file.

The backward adds 2 tests. One that is a numerical check compared to the numpy code, and the other that is a numerical check with the lambda call strategy we use.