Roles Academy

25/07/2024

It is even more difficult deploying an ML Model on a resource constraint devices.

Learn more on how to run ML on Arduino Nano 33: https://youtu.be/Fy_dqeNu2IA

25/07/2024

You can as well run your Image ML Model on the Arduino Nano 33 BLE Sense

PLEASE LIKE AND SUBSCRIBE.Hello there and welcome to this tutorial. I will be walking you through how to build an object recognition model on the Arduino Nan...

15/07/2024

Still on the TinyML Series

PLEEEEAAAASE SUBSCRIBE TO MY CHANNEL, IT MEANS A LOT TO MEWe continue in this tutorial by looking at how to build a Tensorflow Lite model from scratch withou...

13/05/2024

One of the most important board where you can deploy your TinyML workloads is the Arduino Nano 33 BLE sense

09/05/2024

Uploading......................

07/05/2024

If you are looking at starting out in Data Science, check this out for a comprehensive understanding of all the underlying statistics.

Beginners approach to technicalities of Statistics for Data Science

06/05/2024

Tiny Machine Learning involves running machine learning algorithms on low-power microcontrollers or other resource-constrained devices. Enables real-time, low-latency inference on edge devices without relying on cloud connectivity. TinyML can lead to advancements in various fields, including healthcare, agriculture, smart homes, and more.

The advantages of TinyML were summarised by Jeff Bier with the acronym BLERP

I started creating some beginner tutorials on TinyML, you can check it out using this Link: https://youtube.com/playlist?list=PL6vSC9suLAKM2kIRF8O65K5SgQdvzHM69&si=TWaGQnZpDVCN-H-v

28/04/2024

In TinyML, where resources are limited, model size plays a critical role. Traditional machine learning models can be quite large, requiring significant storage space and computational power. This is where quantization comes in.
The Problem: Large Models, Tiny Devices
TinyML applications run on microcontrollers and embedded devices with limited memory (RAM) and processing power (CPU). Large models typically:
• Occupy too much memory: They might not fit on the device entirely or leave insufficient space for other tasks.
• Demand high processing power: Running complex calculations on these models can drain the battery quickly and slow down the device.
The Solution: Quantization - Shrinking the Model for Efficiency
Quantization is a technique used to reduce the size of a machine learning model by reducing the precision of its weights and activations

27/04/2024

https://youtu.be/9sJ3RzJPTbs?si=FaOkh1cWufe2qF4b

Forget the cloud! Discover TinyML, the incredible way to put machine learning directly onto tiny devices. What can you build with this? Find out in this intr...

26/04/2024

Forget the cloud! Discover TinyML, the incredible way to put machine learning directly onto tiny devices. What can you build with this? Find out in this intro to TinyML and get ready to be amazed.

Forget the cloud! Discover TinyML, the incredible way to put machine learning directly onto tiny devices. What can you build with this? Find out in this intr...

16/01/2024

Do you know about the Golden Gate?

15/01/2024

How much did you know about Electric Cars?