Software Open Access

fastmachinelearning/hls4ml: bartsia

vloncar; thesps; Nhan Tran; Ben Kreis; Javier Duarte; jngadiub; Duc Hoang; EJ Kreinar; Adrian Alan Pol; ngpaladi; Dejan Golubovic; Yutaro Iiyama; Zhenbin Wu; Delon; Paolo Cretaro; veyron8800; Petr Zejdl; drankincms

What's new:

  • Streaming IO layer implementations, especially of Convolutional layers, accessed through the config with IOType: io_stream. Scales CNN support to much larger models than previously possible (see arXiv:2101.05108)
  • New documentation and API reference
  • Further optimizations for QKeras / quantization aware training. A 'shift' operation is now used for po2 quantizers
  • Allow redefinition of weights directory for standalone project compilation
  • profiling for PyTorch models

Deprecated:

  • IOType : io_serial is deprecated, and superceded by new IOType: io_stream

Bugfixes:

  • Fix to Initiation Interval and different min/max latency for Strategy: Resource
  • Fix warnings in hls4ml command line script flow
  • Write yml config from Python API - for mixed API / command line flow

Files (1.3 MB)
Name Size
fastmachinelearning/hls4ml-v0.5.0.zip
md5:7cba816e9d2a95deed7588eb3f818fb3
1.3 MB Download
3,739
59
views
downloads
All versions This version
Views 3,73987
Downloads 590
Data volume 227.2 MB0 Bytes
Unique views 3,58783
Unique downloads 560

Share

Cite as