G github.com Show HN: Speed up model inference on CPU with hand crafted layer implementations Kaoken explores the performance of handcrafted layer implementation of common PyTorch layers.The results show that for smaller models, using these "baked" layers enables real time inference without the need for a GPU. ore details in the README. 2 points wanderinglight 1 year ago 2 comments
The results show that for smaller models, using these "baked" layers enables real time inference without the need for a GPU. ore details in the README.