Skip to content

Releases: eki-project/finn-cpp-driver

v1.4

24 Jul 09:36
1f6e4b2

Choose a tag to compare

What's Changed

Full Changelog: v1.3...v1.4

v1.3

13 Jun 10:46
3e67e3d

Choose a tag to compare

What's Changed

Full Changelog: v1.2...v1.3

v1.2

27 Mar 22:44
190cf9a

Choose a tag to compare

What's Changed

Full Changelog: v1.1...v1.2

v1.1

08 May 08:30
c3327c3

Choose a tag to compare

  • Remove hardcoded batch size from kernel execution

  • Implement setBatchSize for complete Stack

  • Remove RingBuffer from Synchronous Inference and add full batch mapping

  • Deduplicate batchsize in basedriver & fix unittests

  • Fix integrationtests

  • Change input kernel code to run concurrrently to output kernel code

  • Optimize inference of lower batch sizes

  • Increase packing performance

  • Further optimize OpenMP

  • Optimize Utils

  • Some small changes

  • Add example data

  • Small Amounts of cleanup

  • Change Driver to run without XRT managed kernels

  • Add more efficient version of execute method

  • Hotfix FPGA bricking

  • Simplify inference interface to speed up inference

  • Update unittest

  • Simplify code

  • Update CMake

  • Fix Release Build CMakeLists

  • Fix wrong old variable names in CMake

  • Fix formatting

  • Change format target

Limitations:

  • Only for XRT based platforms
  • Only tested for Linux based operating systems
  • Currently no support for external weights
  • Asynchronous inference only partially implemented and completely untested -> Will be adressed in future release

v1.0

21 Feb 09:55
21d7ae8

Choose a tag to compare

First release of the C++ driver for FINN.

Features:

  • Most features of the Python driver for FINN
  • A lot faster than the default Python driver
  • Itegrated into FINN

Limitations:

  • Only for XRT based platforms
  • Only tested for Linux based operating systems
  • Currently only partial batch support (Batched inputs are converted to single inputs backend, results in performance penalty for larger batches -> Will be adressed in future release)
  • Asynchronous inference only partially implemented and completely untested -> Will be adressed in future release