Don't have Telegram yet? Try it now!
https://hgpu.org/?p=18759
DNNVM: End-to-End Compiler Leveraging Heterogeneous Optimizations on FPGA-based CNN Accelerators