Releases: plaidml/plaidml
0.7.0 - Final pre-MLIR release
This release contains a number of bug fixes and improvements to performance in the stripe based backends.
It includes full Stripe backends for GPU & CPU for all major targets. Stripe can be used by setting PLAIDML_USE_STRIPE=1 and by ensuring that you pick experimental configs. Stripe backends are only on by default for Intel integrated graphics, as that is the platform with the most advanced Stripe backend.
This release also includes a very high performance experimental CPU backend that achieves nearly state-of-the-art results. It can be activated by setting PLAIDML_STRIPE_JIT=1 PLAIDML_USE_STRIPE=1 and by picking the llvm_cpu device.
We're hoping to get 1.0.0 out the door fast and provide everyone an easy stepping stone into the MLIR world.
0.6.4
This release includes:
- Preview EDSL / plaid2 support
- Initial MLIR integration
- Several workarounds for AMD & iGPU Metal compiler issues
- Fixes for several shape related issues
- New keras backend functions contributed by the community
Other notes:
Stripe based backends should exceed v0 backends for most targets. Enable experimental devices and set USE_STRIPE=1 to test.
0.3.5
See Changelog in README.md
0.3.3rc1
0.3.0
Release Notes:
- Now supports ONNX 1.1.0 as a backend through onnx-plaidml
- Preliminary support for LLVM. Currently only supports CPUs, and only on Linux and macOS. More soon.
- Support for LSTMs & RNNs with static loop sizes, such as examples/imdb_lstm.py (from Keras)
- Adds a shared generic op library in python to make creating frontends easier
- plaidml-keras now uses this library
- Uses plaidml/toolchain for builds
- Building for ARM easy (–-config=linux_arm_32v7)
- Various fixes for bugs (#89)
0.2.0
- Adds new op library and revamps python interface
0.1.3
Fixes several Keras issues
Adds preliminary Windows support
0.1.1
Using pypi is the preferred mechanism for installing PlaidML.
- Adds Python 3 support
- Preview release of macOS