v1.4.2

What's Changed

• Correct the wrong documentation about params in xformer.convert by @LeslieXMOS in #964
• update lib_tflite_micro by @michaelpoluektov in #973
• Fixed condition checking when replacing MaxPool2DOp by @LeslieXMOS in #972
• Bump to python 3.11 by @Allan-xmos in #975

Full Changelog: v1.4.1...v1.4.2

v1.4.1

• Support for allocating input/output tensors together
• Support for allocating input/output tensors in external memory
• Support for paging tensors to external memory
• Support for asynchronous loading of weights from flash
• Switched PyPI release to Jenkins from Github actions

v1.3.1

• Fix Linux release tag

v1.3.0

• Optimized concat, batched softmax, pad, sub, broadcast, along with other ops
• Refactored and improved operator splitting
• Cleaned up loading from flash to support loading weights from flash, tile, or DDR
• Added memory analysis report
• Added more examples
• Updated documentation

v1.2.0

• Added support for int16 ops
• Optimized transpose conv, softmax, slice, along with other ops
• Added denoise audio example
• Updated tensorflow to 2.15
• Updated documentation

v1.1.1

• Fix library inclusion bug

v1.1.0

• Include compiled library to simplify building for xcore
• Better offline memory planning
• Add example apps
• Updated documentation

v1.0.0

• Updated tensorflow to 2.12.0
• Better offline memory planning
• Operation split support
• Multiple models in flash
• Fast flash library
• Bug fixes
• Updated documentation

v0.2.1

• Faster flash
• Multithreading in compiled models
• Add support for more TFlite ops
• Bug fixes

v0.2.0

• Integrated tflite-micro-compiler with xformer
• Multithreading support
• Generalized load from flash
• Optimizations and bugfixes