1.0.0-M1.1
Last updated
Was this helpful?
Last updated
Was this helpful?
A number of bug fixes following the M1 release, thanks to the feedback from the community, allowed us to quickly sort out a few issues. This is a minor bug fix release to address short comings found with M1. Most fixes were related to keras import, the cnn/rnn helpers, and python4j.
Snapshots will also be published every 2 days automatically now to get around sonatype ossrh deletion of snapshots every 3 days. This should increase robustness of the snapshots.
Worked around an issue with github actions pre emptively upgrading visual studio breaking the cuda builds:
Added backwards compatibility for centos 6 via a new linux-x86_64-compat classifier enabling use of older glibcs on centos 7:
A number of bugs were fixed with LSTM and CUDNN:
- avoid shuffle operations on gpu. Pre save data on cpu in mini batches. For more help, please post on the forums at https://community.konduit.ai/
Add batch normalization support for RNNs:
Disable old helpers by default
Minor unit test fixes:
Add keras support for cnn 1d NWHC:
Move the warning about version check to tracing so it stops logging this during normal usage confusing users:
Allow 1d convolutions to accept feed forward as input type:
Remove the old benchmark suite and migrate it to contrib:
Remove old MKLDNNLSTM helper (it never fully functioned anyways):
Fixed an issue with helper reflection ensuring the classes would be loaded properly
Fix minor workspace activation bug:
Fixed compilation error when running anything more than jdk 8 and NIO buffers:
Move logback to be a test dependency for some modules:
Keras model import fixes for GlobalPooling:
Add Eigen op as public ensuring easier use when running eigenvalue decomposition
Fixes minor issue with choice(..) op thanks to
Minor applyScalar typo fix:
Fixed serialization bug with StringToTimeTransform: thanks to community member
Made python4j's python path setting more robust by migrating from set path calls to add path calls:
Fixes bug with numpy import array jvm crashes:
Fixed inconsistent conventions between SameDiffVariable getArr and getArrForName()..