Lava Deep Learning 0.3.0

[mgkwill]

The lava-dl library version 0.3.0 now enables inference for trained spiking networks seamlessly on CPU or Loihi 2 backends and can leverage Loihi 2’s convolutional network compression and graded spike features for improved memory usage and performance.

New Features and Improvements

• Added Loihi 2 support in lava-dl NetX utilizing Loihi 2 convolution support and graded spikes (PR Changes to lava-dl to reflect api changes in lava 0.4.0 #88, Loihi Tutorials #107).
• Added a tutorial demonstrating PilotNet application running on Intel Loihi 2 (PR Loihi Tutorials #107).
• Added accelerated training of recurrent topologies in lava-dl SLAYER (PR Recurrent mechanic #103)
• Added Transposed Convolution and Unpool support in lava-dl SLAYER (PR Added ConvT and Unpool block for neurons #80)

Bug Fixes and Other Changes

• Improved lava-dl SLAYER codebase with bugfixes and additional documentation (PR Slayer fixes #78, fix bug of block.AbstractInput #105)

Breaking Changes

• No breaking changes in this release

Known Issues

• No known issues in this release

What's Changed

• Remove unnecessary imports by @Tobias-Fischer in Remove unnecessary imports #56
• Update mnist.py by @uslumt in Update mnist.py #71
• Updated notebooks with new hyperparameters and typo fixes by @bamsumit in Updated notebooks with new hyperparameters and typo fixes #76
• Add pilotnet integration tests by @mgkwill in Add pilotnet integration tests #79
• Slayer fixes by @bamsumit in Slayer fixes #78
• Changes to lava-dl to reflect api changes in lava 0.4.0 by @bamsumit in Changes to lava-dl to reflect api changes in lava 0.4.0 #88
• Added ConvT and Unpool block for neurons by @alexggener in Added ConvT and Unpool block for neurons #80
• Update ci-build.yml, Remove redundant poetry updates by @mgkwill in Update ci-build.yml, Remove redundant poetry updates #89
• Recurrent mechanic by @timcheck in Recurrent mechanic #103
• Bump nbconvert from 6.5.0 to 6.5.1 by @dependabot in Bump nbconvert from 6.5.0 to 6.5.1 #93
• fix bug of block.AbstractInput by @fangwei123456 in fix bug of block.AbstractInput #105
• Loihi Tutorials by @bamsumit in Loihi Tutorials #107
• Add conda install instructions with intel-numpy by @mgkwill in Add conda install instructions with intel-numpy #91
• Version 0.3.0 by @mgkwill in Version 0.3.0 #109

New Contributors

• @Tobias-Fischer made their first contribution in Remove unnecessary imports #56
• @uslumt made their first contribution in Update mnist.py #71
• @alexggener made their first contribution in Added ConvT and Unpool block for neurons #80
• @timcheck made their first contribution in Recurrent mechanic #103
• @dependabot made their first contribution in Bump nbconvert from 6.5.0 to 6.5.1 #93
• @fangwei123456 made their first contribution in fix bug of block.AbstractInput #105

Full Changelog: v0.2.0...v0.3.0

---

This discussion was created from the release Lava Deep Learning 0.3.0.

[Dola47]

@mgkwill @bamsumit

May I ask how am I supposed to cite laval-dl?

Thanks.

[bamsumit]

@Dola47 we do not have an official publication for lava-dl. You can always cite the github link. Most of lava-dl functionality is based on SLAYER paper. You can cite that if you are using lava.lib.dl.slayer. And we always appreciate github stars.