Blame - release_notes.txt - ml/ethos-u/ml-embedded-evaluation-kit

blob: e70332bb5a28fd7aa5268852528c56ae11ea1995 [file] [log] [blame]

Richard Burton	49482d5	2023-11-30 11:38:45 +0000	[diff] [blame]	1	Changes in 23.11
				2	* Support for 23.11 NPU components and dependencies (core-driver, core-platform, Vela 3.10.0, CMSIS, CMSIS-NN and TensorFlow Lite Micro).
				3	* Increase minimum requirements of Python to 3.10
				4	* Update Python files to conform to Pylint
				5	* Adding support for Arm Cortex-M85 with GCC
				6	* Updates to documentation
				7
Nina Drozd	05c1c5a	2023-08-22 12:41:55 +0100	[diff] [blame]	8	Changes in 23.08
				9	* Support for 23.08 NPU components and dependencies (core-driver, core-platform, Vela 3.9.0, CMSIS, CMSIS-NN and TensorFlow Lite Micro).
				10	* Update Python package requirements for compatibility reasons.
Nina Drozd	9263b51	2023-09-05 17:38:30 +0100	[diff] [blame]	11	* Increase minimum requirements of Armclang to 6.19
Nina Drozd	05c1c5a	2023-08-22 12:41:55 +0100	[diff] [blame]	12
Richard Burton	df88c5d	2023-05-30 17:40:39 +0100	[diff] [blame]	13	Changes in 23.05
				14	* Support for 23.05 NPU components and dependencies (core-driver, core-platform, Vela 3.8.0, CMSIS, CMSIS-NN and TensorFlow Lite Micro).
				15	* Minor updates to CMSIS-pack dependencies.
				16	* Update Python package requirements for compatibility reasons.
				17	* Increase minimum requirements of Python to 3.9
				18
Kshitij Sisodia	26bc923	2023-03-10 16:33:23 +0000	[diff] [blame]	19	Changes in 23.02
				20	* Support for 23.02 NPU components and dependencies (core-driver, core-platform, Vela 3.7.0, CMSIS, CMSIS-DSP, CMSIS-NN and TensorFlow Lite Micro).
				21	* Improvement to PMU counters - reducing cache maintenance burden for Arm Cortex-M55 CPUs when NPU is used.
				22	* Inclusive language updates.
				23	* Documentation updates including adding links to AN555 resources.
				24	* Updates to Python requirements file for virtual environment.
				25
Kshitij Sisodia	7ac0a3a	2022-11-09 14:12:08 +0000	[diff] [blame]	26	Changes in 22.11
Richard Burton	71f282e	2022-12-01 12:31:23 +0000	[diff] [blame]	27	* Support for 22.11 NPU components and dependencies (core-driver, core-platform, Vela 3.6.0, CMSIS, TensorFlow Lite Micro)
				28	* CMSIS-NN is added as a new dependency due to it moving out of the core CMSIS repository.
Kshitij Sisodia	7ac0a3a	2022-11-09 14:12:08 +0000	[diff] [blame]	29	* Updates for Arm® Corstone™-310 platform.
				30	* Formal support for CMake presets.
				31	* Support for the new Arm Compiler 6.19
				32	* Published CMSIS-Pack under `resources` and related updates.
				33	* Bug fix for building with NPU disabled.
				34
Kshitij Sisodia	01334f9	2022-08-12 11:53:45 +0100	[diff] [blame]	35	Changes in 22.08
Richard Burton	71f282e	2022-12-01 12:31:23 +0000	[diff] [blame]	36	* Support for 22.08 NPU component and dependencies (core-driver, core-platform, Vela 3.5.0, CMSIS, TensorFlow Lite Micro).
Kshitij Sisodia	01334f9	2022-08-12 11:53:45 +0100	[diff] [blame]	37	* Formal support for Arm® Corstone-310 Arm Virtual Hardware and FPGA implementations.
				38	* Added CMake presets.
				39	* Bug fix for Arm® Ethos-U65 NPU's default AXI burst length.
				40	* Provision to allow platform-agnostic use case API to be packaged as CMSIS pack.
				41	* Documentation updates.
				42
Richard Burton	5dd1838	2022-06-01 10:26:46 +0100	[diff] [blame]	43	Changes in 22.05
Richard Burton	71f282e	2022-12-01 12:31:23 +0000	[diff] [blame]	44	* Support for 22.05 NPU component and dependencies (core-driver, core-platform, Vela 3.4.0, CMSIS, TensorFlow Lite Micro).
Richard Burton	5dd1838	2022-06-01 10:26:46 +0100	[diff] [blame]	45	* Restructuring of repository sources to allow generation of CMSIS packs.
Kshitij Sisodia	d088e52	2022-08-17 09:35:09 +0100	[diff] [blame]	46	* Added support for Arm Cortex-M85, Armclang 6.18 is required
Richard Burton	5dd1838	2022-06-01 10:26:46 +0100	[diff] [blame]	47	* Reduce minimum requirements of Python to 3.7 to align with Vela 3.4.0.
				48	* Increase minimum requirements of CMake to 3.21.0
				49	* Documentation updates.
				50	* Various bug fixes.
				51
Richard Burton	b4123fd	2022-03-04 09:19:09 +0000	[diff] [blame]	52	Changes in 22.02
Richard Burton	71f282e	2022-12-01 12:31:23 +0000	[diff] [blame]	53	* Support for 22.02 NPU component and dependencies (core-driver, core-platform, Vela 3.3.0, CMSIS, TensorFlow Lite Micro).
Richard Burton	b4123fd	2022-03-04 09:19:09 +0000	[diff] [blame]	54	* Added Object Detection use case.
				55	* Replace DSCNN with MicroNet for KWS and KWS_ASR use cases.
Kshitij Sisodia	d088e52	2022-08-17 09:35:09 +0100	[diff] [blame]	56	* Increase minimum requirements of CMake to 3.15.6, Armclang to 6.16 and Python to 3.8
Richard Burton	b4123fd	2022-03-04 09:19:09 +0000	[diff] [blame]	57	* Initial restructuring of repository sources.
				58	* Documentation updates.
				59	* Various minor bug fixes.
				60
Isabella Gottardi	118f73e	2021-09-16 17:54:35 +0100	[diff] [blame]	61	Changes in 21.11
Richard Burton	71f282e	2022-12-01 12:31:23 +0000	[diff] [blame]	62	* Support for 21.11 NPU component and dependencies (core-software, core-driver, Vela 3.2.0, CMSIS, TensorFlow Lite Micro).
Isabella Gottardi	118f73e	2021-09-16 17:54:35 +0100	[diff] [blame]	63	* Added dynamic load support for FVP for inference runner use-case.
Isabella Gottardi	84f702d	2021-11-11 17:06:12 +0000	[diff] [blame]	64	* Added support for different memory modes: Shared_Sram, Dedicated_Sram and Sram_Only.
				65	* Added Noise Reduction use case.
				66	* Updated support for Arm GNU Embedded Toolchain 10.3-2021.07 and Arm Compiler 6.17.
Isabella Gottardi	118f73e	2021-09-16 17:54:35 +0100	[diff] [blame]	67	* Documentation updates.
Kshitij Sisodia	661959c	2021-11-24 10:39:52 +0000	[diff] [blame]	68	* Changes to support AN552 design - the new Arm® Corstone™-300 implementation.
Isabella Gottardi	118f73e	2021-09-16 17:54:35 +0100	[diff] [blame]	69
Isabella Gottardi	c5d8bda	2021-07-21 10:35:08 +0100	[diff] [blame]	70	Changes in 21.08
Richard Burton	71f282e	2022-12-01 12:31:23 +0000	[diff] [blame]	71	* Support for 21.05 NPU component (core-software, core-driver, Vela 3.0.0).
Isabella Gottardi	c5d8bda	2021-07-21 10:35:08 +0100	[diff] [blame]	72	* TensorFlow submodule changed to https://github.com/tensorflow/tflite-micro (tested with TensorFlow Lite Micro commit hash: f510d38d0eaa3195ce3af66e3f32648740f08afb).
				73	* Image classification model changed (from uint8 to int8 datatype).
Richard Burton	71f282e	2022-12-01 12:31:23 +0000	[diff] [blame]	74	* Documentation updates, added support to Corstone-300 + Arm® Ethos-U65.
Isabella Gottardi	c5d8bda	2021-07-21 10:35:08 +0100	[diff] [blame]	75	* Various scripts improvements.
Isabella Gottardi	118f73e	2021-09-16 17:54:35 +0100	[diff] [blame]	76	* Added Visual Wake Word use case.
Isabella Gottardi	c5d8bda	2021-07-21 10:35:08 +0100	[diff] [blame]	77
Isabella Gottardi	2181d0a	2021-04-07 09:27:38 +0100	[diff] [blame]	78	Changes in 21.05
				79	* Added script to download and optimize default models.
				80	* Added script to run default build flow.
				81	* Added a model for Anomaly Detection use case.
Isabella Gottardi	c5d8bda	2021-07-21 10:35:08 +0100	[diff] [blame]	82	* Added support for build with Arm GNU Embedded Toolchain (10.2.1).
				83	* Deprecated support for target subsystem SSE-200.
Isabella Gottardi	2181d0a	2021-04-07 09:27:38 +0100	[diff] [blame]	84
alexander	3c79893	2021-03-26 21:42:19 +0000	[diff] [blame]	85	Changes in 21.03
Isabella Gottardi	c5d8bda	2021-07-21 10:35:08 +0100	[diff] [blame]	86	* Simple platform support added.
				87	* Model conditioning examples added.
				88	* Documentation updated.
				89	* Build changed to use sources of the dependency libraries.
				90	* Tests for native platform added.
				91	* Anomaly detection use-case added.
alexander	3c79893	2021-03-26 21:42:19 +0000	[diff] [blame]	92
				93	Changes in 20.11
				94	* SSE-200 and SSE-300 system support was added.
Richard Burton	71f282e	2022-12-01 12:31:23 +0000	[diff] [blame]	95	* Support for simple fixed virtual platform for Arm® Ethos-U55 and Arm® Cortex-M55 removed.
alexander	3c79893	2021-03-26 21:42:19 +0000	[diff] [blame]	96	* Build cmake parameters changed: TARGET_SUBSYSTEM was added, TARGET_PLATFORM accepted values were changed.
				97	* Models with multiple output tensors support was added.
				98	* Generic inference runner use-case added.
				99	* ASR triggered by KWS added in the same use case (kws_asr). This also shows how to re-use tensor arena with two models using the same memory pool.
				100
				101	Changes in 20.09 release:
				102	* Support for TensorFlow Lite Micro version > 2.3.0 (tested with TensorFlow Lite Micro 2.4.0 commit hash: 5bbb8a2bd1def6865b1510175a3da5fd12387e10)
				103	* Added speech recognition use case example.
Richard Burton	71f282e	2022-12-01 12:31:23 +0000	[diff] [blame]	104	* Updated NPU Fastmodel version to r0p2-00eac0-rc4
alexander	3c79893	2021-03-26 21:42:19 +0000	[diff] [blame]	105
				106	Changes in 20.08 release:
				107	* Added keyword spotting use case example.
				108	* Added person detection use case example.
				109	Known issues:
Conor Kennedy	5cf8e74	2023-02-13 10:50:40 +0000	[diff] [blame]	110	* telnet connection to FastModel environment may stop responding after some period of inactivity.
alexander	3c79893	2021-03-26 21:42:19 +0000	[diff] [blame]	111
				112	Changes in 20.05 release:
				113	* FastModel environment was built with FastModel Tools v11.10.22.
				114	* Mps3 FPGA build support was added.
				115	* Configurable timing-adaptor.
				116	* Added Active and Idle cycle counts for NPU and CPU profiling report.
				117	* Source code structure and build scripts refactored to support multiple ML use-cases.
Richard Burton	71f282e	2022-12-01 12:31:23 +0000	[diff] [blame]	118	* Used EAC NPU software model and drivers.
alexander	3c79893	2021-03-26 21:42:19 +0000	[diff] [blame]	119	* Windows support for build scripts.
				120	Known issues:
Conor Kennedy	5cf8e74	2023-02-13 10:50:40 +0000	[diff] [blame]	121	* telnet connection to FastModel environment may stop responding after some period of inactivity.