From 1716efd0b35889b580276e27c8b6f661c9858cd0 Mon Sep 17 00:00:00 2001 From: Isabella Gottardi Date: Tue, 8 Mar 2022 15:27:49 +0000 Subject: MLECO-3006: Fixing some minor errors in documentation Change-Id: I24cd544780f46fcec8f154b440f7bb959c20a459 Signed-off-by: Isabella Gottardi --- docs/documentation.md | 32 +++++++++++++++++++------------- 1 file changed, 19 insertions(+), 13 deletions(-) (limited to 'docs/documentation.md') diff --git a/docs/documentation.md b/docs/documentation.md index f911cff..9a00cc4 100644 --- a/docs/documentation.md +++ b/docs/documentation.md @@ -203,10 +203,12 @@ What these folders contain: through `CMSIS_SRC_PATH` variable. The static library is used by platform code. -- `components` directory contains drivers code for different devices used in platforms. Such as UART, LCD and others. - A platform can include those as sources in a build to enable usage of corresponding HW devices. Most of the use-cases - use UART and LCD, thus if you want to run default ML use-cases on a custom platform, you will have to add - implementation for your devices here (or re-use existing code if it is compatible with your platform). +- `components` directory contains drivers for different modules that can be reused for different platforms. + These contain common functions for Arm Ethos-U NPU initialization, timing adapter block helpers and others. + Each component produces a static library that could potentially be linked into the platform library to enable + usage of corresponding modules from the platform sources. For example, most of the use-cases use NPU and + timing adapter initialization. If you want to run default ML use-cases on a custom platform, you could re-use + existing code from this directory provided it is compatible with your platform. - `platform/mps3`\ `platform/simple`: @@ -228,18 +230,22 @@ What these folders contain: Native profile allows to build application to be executed on a build machine, i.e. x86. It bypasses and stubs platform devices replacing them with standard C or C++ library calls. -- `platforms/bare-metal/bsp/mem_layout`: Contains the platform-specific linker scripts. - ## Models and resources -The models used in the use-cases implemented in this project can be downloaded from: [Arm ML-Zoo](https://github.com/ARM-software/ML-zoo). +The models used in the use-cases implemented in this project can be downloaded from: + +- [Arm ML-Zoo](https://github.com/ARM-software/ML-zoo) ( [Apache 2.0 License](https://github.com/ARM-software/ML-zoo/blob/master/LICENSE) ) + + - [Mobilenet V2](https://github.com/ARM-software/ML-zoo/tree/e0aa361b03c738047b9147d1a50e3f2dcb13dbcb/models/image_classification/mobilenet_v2_1.0_224/tflite_int8) + - [MicroNet for Keyword Spotting](https://github.com/ARM-software/ML-zoo/tree/9f506fe52b39df545f0e6c5ff9223f671bc5ae00/models/keyword_spotting/micronet_medium/tflite_int8) + - [Wav2Letter](https://github.com/ARM-software/ML-zoo/tree/1a92aa08c0de49a7304e0a7f3f59df6f4fd33ac8/models/speech_recognition/wav2letter/tflite_pruned_int8) + - [MicroNet for Anomaly Detection](https://github.com/ARM-software/ML-zoo/tree/7c32b097f7d94aae2cd0b98a8ed5a3ba81e66b18/models/anomaly_detection/micronet_medium/tflite_int8) + - [MicroNet for Visual Wake Word](https://github.com/ARM-software/ML-zoo/raw/7dd3b16bb84007daf88be8648983c07f3eb21140/models/visual_wake_words/micronet_vww4/tflite_int8/vww4_128_128_INT8.tflite) + - [RNNoise](https://github.com/ARM-software/ML-zoo/raw/a061600058097a2785d6f1f7785e5a2d2a142955/models/noise_suppression/RNNoise/tflite_int8/rnnoise_INT8.tflite) + +- [Emza Visual Sense ModelZoo](https://github.com/emza-vs/ModelZoo) ( [Apache 2.0 License](https://github.com/emza-vs/ModelZoo/blob/v1.0/LICENSE) ) -- [Mobilenet V2](https://github.com/ARM-software/ML-zoo/tree/e0aa361b03c738047b9147d1a50e3f2dcb13dbcb/models/image_classification/mobilenet_v2_1.0_224/tflite_int8) -- [MicroNet for Keyword Spotting](https://github.com/ARM-software/ML-zoo/tree/9f506fe52b39df545f0e6c5ff9223f671bc5ae00/models/keyword_spotting/micronet_medium/tflite_int8) -- [Wav2Letter](https://github.com/ARM-software/ML-zoo/tree/1a92aa08c0de49a7304e0a7f3f59df6f4fd33ac8/models/speech_recognition/wav2letter/tflite_pruned_int8) -- [MicroNet for Anomaly Detection](https://github.com/ARM-software/ML-zoo/tree/7c32b097f7d94aae2cd0b98a8ed5a3ba81e66b18/models/anomaly_detection/micronet_medium/tflite_int8) -- [MicroNet for Visual Wake Word](https://github.com/ARM-software/ML-zoo/raw/7dd3b16bb84007daf88be8648983c07f3eb21140/models/visual_wake_words/micronet_vww4/tflite_int8/vww4_128_128_INT8.tflite) -- [RNNoise](https://github.com/ARM-software/ML-zoo/raw/a061600058097a2785d6f1f7785e5a2d2a142955/models/noise_suppression/RNNoise/tflite_int8/rnnoise_INT8.tflite) + - [YOLO Fastest](https://github.com/emza-vs/ModelZoo/blob/v1.0/object_detection/yolo-fastest_192_face_v4.tflite) When using *Ethos-U* NPU backend, Vela compiler optimizes the the NN model. However, if not and it is supported by TensorFlow Lite Micro, then it falls back on the CPU and execute. -- cgit v1.2.1