Verified Commit 0ed5f0ce authored by Gergely Nagy's avatar Gergely Nagy Committed by Vincent Coubard
Browse files

mlia: Update READMEs for MLIA

Signed-off-by: Gergely Nagy's avatarGergely Nagy <>
parent a16679f6
......@@ -472,6 +472,7 @@ Execute the following script located in the application repository.
- `kws/source/main_ns.c`: Entry point of the kws application.
- `kws/source/blinky_task.c`: Blinky/UX thread of the application.
- `kws/source/ml_interface.c`: Interface between the virtual streaming solution and tensor flow.
- `mlia`: Integration the ML Inference Advisor, using a simple wrapper to install and run the tool on given models.
# ML Model Replacement
# ML Inference Advisor
## Introduction
This tool is used to help AI developers design and optimize neural network
models for efficient inference on Arm targets by enabling performance analysis
and providing actionable advice early in the model development cycle. The final
advice can cover the operator list, performance analysis and suggestions for
model inference run on certain hardware before/after applying model optimization
(e.g. pruning, clustering, etc.).
## Backends
The ML Inference Advisor is designed to support multiple performance
estimators (backends) to generate performance analysis for individual
types of hardware.
MLIA currently supports the following backend platforms:
* Corstone-300 - see
* Corstone-310 - see
## Setup
To set up MLIA with the [Arm IoT Total Solutions]( environment,
the recommended way is to use the `` frontend script:
../ build mlia
This command installs the latest released version of MLIA along with its dependencies, and sets up the Python
virtual environment for the subsequent commands.
## Usage via the frontend script
Once the environemnt is set up, you can use the frontend script, `` for a quick MLIA test drive:
../ run mlia
This will run MLIA operator command on a bundled quantized DS-CNN-L model, which reports advice.
## Using MLIA CLI directly
In order to access the full feature set of the ML Inference Advisor tool, you can use the CLI directly as follows.
First you need to activate the virtual environment set up by the `build` command above.
In the top level of the `keyword` repository, type:
source build/mlia/venv/bin/activate
After this you can run `mlia`:
mlia --help
# Reference
For more details, please refer to official documentation:
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment