[Announcement] Apache TVM (incubating) v0.6.0 Release

Hi all,

The Apache TVM (incubating) community is happy to announce Apache TVM (incubating) version 0.6.0!

Apache TVM (incubating) is an open deep learning compiler stack for CPUs, GPUs, and specialized accelerators. It aims to close the gap between the productivity-focused deep learning frameworks, and the performance- or efficiency-oriented hardware backends.

0.6.0 is a major release incorporating important features including high-level Relay IR, Relay Virtual Machine, training support, quantization, TSim, MicroTVM, Rust support, etc. It also contains a count of operator support and performance improvement.

A full list of the changes in this release can be found in the release notes:

A link to the download can be found here: https://dist.apache.org/repos/dist/release/incubator/tvm/tvm-v0.6.0/

The Release Tag:

TVM Resources

Best regards, Apache TVM (incubating) Team


DISCLAIMER:

Apache TVM (incubating) is an effort undergoing incubation at The Apache Software Foundation (ASF), sponsored by the Apache Incubator PMC.

Incubation is required of all newly accepted projects until a further review indicates that the infrastructure, communications, and decision making process have stabilized in a manner consistent with other successful ASF projects.

11 Likes

Can we include “Fix search path for libtvm_topi.so” https://github.com/apache/incubator-tvm/commit/6ef2418b3e4c962448a2cb00968f484bc35888c3 into v0.6? Otherwise people can not import relay from tvm. They need to use LD_LIBRARY_PATH and point to where libtvm_topi.so is. https://github.com/apache/incubator-tvm/issues/4466

This is something that we can cherry pick and send to the v0.6 branch. We can do a followup minor release later. In the meanwhile, the normal PYTHONPATH or the LD_LIBRARY_PATH path approach works for v0.6.0

Now, Are there any optimizer in tvm, like SGD, Adam?