element14 Community
element14 Community
    Register Log In
  • Site
  • Search
  • Log In Register
  • Members
    Members
    • Benefits of Membership
    • Achievement Levels
    • Members Area
    • Personal Blogs
    • Feedback and Support
    • What's New on element14
  • Learn
    Learn
    • Learning Center
    • eBooks
    • STEM Academy
    • Webinars, Training and Events
    • More
  • Technologies
    Technologies
    • 3D Printing
    • FPGA
    • Industrial Automation
    • Internet of Things
    • Power & Energy
    • Sensors
    • More
  • Challenges & Projects
    Challenges & Projects
    • Design Challenges
    • element14 presents
    • Project14
    • Arduino Projects
    • Raspberry Pi Projects
    • More
  • Products
    Products
    • Arduino
    • Dev Tools
    • Manufacturers
    • Raspberry Pi
    • RoadTests & Reviews
    • Avnet Boards Community
    • More
  • Store
    Store
    • Visit Your Store
    • Choose Another Store
      • Europe
      •  Austria (German)
      •  Belgium (Dutch, French)
      •  Bulgaria (Bulgarian)
      •  Czech Republic (Czech)
      •  Denmark (Danish)
      •  Estonia (Estonian)
      •  Finland (Finnish)
      •  France (French)
      •  Germany (German)
      •  Hungary (Hungarian)
      •  Ireland
      •  Israel
      •  Italy (Italian)
      •  Latvia (Latvian)
      •  
      •  Lithuania (Lithuanian)
      •  Netherlands (Dutch)
      •  Norway (Norwegian)
      •  Poland (Polish)
      •  Portugal (Portuguese)
      •  Romania (Romanian)
      •  Russia (Russian)
      •  Slovakia (Slovak)
      •  Slovenia (Slovenian)
      •  Spain (Spanish)
      •  Sweden (Swedish)
      •  Switzerland(German, French)
      •  Turkey (Turkish)
      •  United Kingdom
      • Asia Pacific
      •  Australia
      •  China
      •  Hong Kong
      •  India
      •  Korea (Korean)
      •  Malaysia
      •  New Zealand
      •  Philippines
      •  Singapore
      •  Taiwan
      •  Thailand (Thai)
      • Americas
      •  Brazil (Portuguese)
      •  Canada
      •  Mexico (Spanish)
      •  United States
      Can't find the country/region you're looking for? Visit our export site or find a local distributor.
  • Translate
  • Profile
Pi-Fest
  • Challenges & Projects
  • Design Challenges
  • Pi-Fest
  • More
  • Cancel
Pi-Fest
Blog Songspire - TinyML
  • Blog
  • Forum
  • Documents
  • Leaderboard
  • Mentions
  • Sub-Groups
  • Tags
  • More
  • Cancel
  • New
Blog Post Actions
  • Subscribe by email
  • More
  • Cancel
  • Share
  • Subscribe by email
  • More
  • Cancel
Group Actions
  • Group RSS
  • More
  • Cancel
Engagement
  • Author Author: feiticeir0
  • Date Created: 22 Jun 2022 11:21 AM Date Created
  • Views 72 views
  • Likes 5 likes
  • Comments 2 comments
  • raspberry pi pico
  • tinyml
  • machine learning
  • pi-fest
  • pi-fest songspire
Related
Recommended

Songspire - TinyML

feiticeir0
feiticeir0
22 Jun 2022
Songspire - TinyML

Hi everyone ! Hope you are all well.

TinyML

TinyML referes to Tiny Machine Learning.  It is a new and growing field of Machine Learning applied to resource-constrained hardware - such as an RP2040 like the PICO.

When speaking of resource constrained hardware, it's referring  to the limited memory and processing power available on a microcontroller compared with a server and GPU.

TinyML allows us to interpret data and do some inference on devices powered by coin cells . This devices are also known as edge devices - because instead of uploading data to the cloud, the tinyML tools are available locally.

This brings some advantages:

  • Edge devices reduce the costs when talking of cellular connectivity, when there are no WiFi networks available. The data can be only uploaded when something is detected, instead of having to send each packet of data to be analyzed in the cloud.
  • Battery life is extended because the device can remain in a sleep state until an event is detected.

Here's a typical usage with TinyML, where the inference is done in the edge device.

In the first example, all data is sent to the cloud, where data is analyzed (and inferred) and results are sent back to the device.

At the Edge, inference is done at the device and the results are optionally sent to the cloud.

Applications

This is an outgoing field with a lot of study and development still being actively done. Still, what can we do with TinyML

  • Visual Inspection
  • Predictive maintenance (using accelerometers to detect vibrations)
  • Monitoring flora/fauna: Using sound to detect vehicles or hacksaws to prevent poachers
  • Image classification

TinyML and Machine Learning

TinyML is a subset of Machine Learning, which in turn is a subset of Artificial Intelligence. ML is used to train a machine to learn from data and make predictions or decisions on its own.

ML has several toolsets, being deep learning one of them.  This is a technique that is modeled after the human brain, where information passes through a network of layers aimed at learning patterns.

Each layer contains several nodes, called neurons. They are arranged into input, output and the ones in the middle.  This is called a neural network.  Each node has an associated weight.

Neural Networks typically require vast amounts of processing power and usually run on servers. TinyML's tools allow to run a modified neural network on a microcontroller.

Deploying on a Microcontroller

One of the most critical aspects to consider when deploying Deep Learning networks for TinyML is the model size, generally defined as the memory required. Since the tiny platforms have limited physical memory, the model must be compact to fit the target devices.

However, the memory is not the only challenge when deploying a model on a microcontroller.

The trained models commonly employs arithmetic operations with floating-point precision, but, the CPUs on Microcontrollers usually don't have the hardware acceleration for it.

So, quantization is an indispensable technique to overcome it.

Quantization

Quantization is the process by which the precision (bit size) of the model's weights and biases, so that the model takes less memory, runs faster and requires less power - with a minimal hit to accuracy.

The widely adopted technique for microcontrollers applies the quantization post-training by converting the 32-bit floating point weights to 8-bit integer values. This brings the model 4x smaller and a latency improvement .

On the next post, will talk about the stages involved on creating a model for TinyML and deploying it on a device. In this post, we will also talk about the audio classification - preparation of the files.

References

https://news.mit.edu/2017/explained-neural-networks-deep-learning-0414

Raspberry PI Pico DIY Workshop, Sai Yamanoor, Srihari Yamanoor , Pack Publishing, 2022

TinyML - Machine Learning with TensorFlow Lite on Arduino and Ultra-Lowe-Power Microcontrollers, Pete Warden & Daniel Situnayake, O' Reilly, 2019

TinyML Cookbook, Gian Marco Iodice, Pack Publishing, 2022

%3CmxGraphModel%3E%3Croot%3E%3CmxCell%20id%3D%220%22%2F%3E%3CmxCell%20id%3D%221%22%20parent%3D%220%22%2F%3E%3CmxCell%20id%3D%222%22%20value%3D%22%22%20style%3D%22ellipse%3BwhiteSpace%3Dwrap%3Bhtml%3D1%3Baspect%3Dfixed%3BfillColor%3D%23e3c800%3BfontColor%3D%23000000%3BstrokeColor%3D%23B09500%3B%22%20vertex%3D%221%22%20parent%3D%221%22%3E%3CmxGeometry%20x%3D%2290%22%20y%3D%22270%22%20width%3D%2250%22%20height%3D%2250%22%20as%3D%22geometry%22%2F%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%223%22%20value%3D%22%22%20style%3D%22ellipse%3BwhiteSpace%3Dwrap%3Bhtml%3D1%3Baspect%3Dfixed%3BfillColor%3D%23e3c800%3BfontColor%3D%23000000%3BstrokeColor%3D%23B09500%3B%22%20vertex%3D%221%22%20parent%3D%221%22%3E%3CmxGeometry%20x%3D%2290%22%20y%3D%22340%22%20width%3D%2250%22%20height%3D%2250%22%20as%3D%22geometry%22%2F%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%224%22%20value%3D%22%22%20style%3D%22ellipse%3BwhiteSpace%3Dwrap%3Bhtml%3D1%3Baspect%3Dfixed%3BfillColor%3D%23d0cee2%3BstrokeColor%3D%2356517e%3B%22%20vertex%3D%221%22%20parent%3D%221%22%3E%3CmxGeometry%20x%3D%22210%22%20y%3D%22200%22%20width%3D%2250%22%20height%3D%2250%22%20as%3D%22geometry%22%2F%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%225%22%20value%3D%22%22%20style%3D%22ellipse%3BwhiteSpace%3Dwrap%3Bhtml%3D1%3Baspect%3Dfixed%3BfillColor%3D%23d0cee2%3BstrokeColor%3D%2356517e%3B%22%20vertex%3D%221%22%20parent%3D%221%22%3E%3CmxGeometry%20x%3D%22210%22%20y%3D%22270%22%20width%3D%2250%22%20height%3D%2250%22%20as%3D%22geometry%22%2F%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%226%22%20value%3D%22%22%20style%3D%22ellipse%3BwhiteSpace%3Dwrap%3Bhtml%3D1%3Baspect%3Dfixed%3BfillColor%3D%23d0cee2%3BstrokeColor%3D%2356517e%3B%22%20vertex%3D%221%22%20parent%3D%221%22%3E%3CmxGeometry%20x%3D%22210%22%20y%3D%22340%22%20width%3D%2250%22%20height%3D%2250%22%20as%3D%22geometry%22%2F%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%227%22%20value%3D%22%22%20style%3D%22ellipse%3BwhiteSpace%3Dwrap%3Bhtml%3D1%3Baspect%3Dfixed%3BfillColor%3D%23d0cee2%3BstrokeColor%3D%2356517e%3B%22%20vertex%3D%221%22%20parent%3D%221%22%3E%3CmxGeometry%20x%3D%22210%22%20y%3D%22410%22%20width%3D%2250%22%20height%3D%2250%22%20as%3D%22geometry%22%2F%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%228%22%20value%3D%22%22%20style%3D%22ellipse%3BwhiteSpace%3Dwrap%3Bhtml%3D1%3Baspect%3Dfixed%3BfillColor%3D%23b1ddf0%3BstrokeColor%3D%2310739e%3B%22%20vertex%3D%221%22%20parent%3D%221%22%3E%3CmxGeometry%20x%3D%22320%22%20y%3D%22310%22%20width%3D%2250%22%20height%3D%2250%22%20as%3D%22geometry%22%2F%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%229%22%20value%3D%22%22%20style%3D%22endArrow%3Dclassic%3Bhtml%3D1%3Brounded%3D0%3BstrokeColor%3D%23FF3333%3BexitX%3D1%3BexitY%3D0.5%3BexitDx%3D0%3BexitDy%3D0%3BentryX%3D0%3BentryY%3D0.5%3BentryDx%3D0%3BentryDy%3D0%3B%22%20edge%3D%221%22%20source%3D%222%22%20target%3D%224%22%20parent%3D%221%22%3E%3CmxGeometry%20width%3D%2250%22%20height%3D%2250%22%20relative%3D%221%22%20as%3D%22geometry%22%3E%3CmxPoint%20x%3D%22130%22%20y%3D%22250%22%20as%3D%22sourcePoint%22%2F%3E%3CmxPoint%20x%3D%22180%22%20y%3D%22200%22%20as%3D%22targetPoint%22%2F%3E%3C%2FmxGeometry%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%2210%22%20value%3D%22%22%20style%3D%22endArrow%3Dclassic%3Bhtml%3D1%3Brounded%3D0%3BstrokeColor%3D%23FF3333%3BexitX%3D1%3BexitY%3D0.5%3BexitDx%3D0%3BexitDy%3D0%3BentryX%3D0%3BentryY%3D0.5%3BentryDx%3D0%3BentryDy%3D0%3B%22%20edge%3D%221%22%20source%3D%222%22%20target%3D%225%22%20parent%3D%221%22%3E%3CmxGeometry%20width%3D%2250%22%20height%3D%2250%22%20relative%3D%221%22%20as%3D%22geometry%22%3E%3CmxPoint%20x%3D%22150%22%20y%3D%22305%22%20as%3D%22sourcePoint%22%2F%3E%3CmxPoint%20x%3D%22220%22%20y%3D%22235%22%20as%3D%22targetPoint%22%2F%3E%3C%2FmxGeometry%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%2211%22%20value%3D%22%22%20style%3D%22endArrow%3Dclassic%3Bhtml%3D1%3Brounded%3D0%3BstrokeColor%3D%23FF3333%3BexitX%3D1%3BexitY%3D0.5%3BexitDx%3D0%3BexitDy%3D0%3BentryX%3D-0.036%3BentryY%3D0.446%3BentryDx%3D0%3BentryDy%3D0%3BentryPerimeter%3D0%3B%22%20edge%3D%221%22%20source%3D%222%22%20target%3D%226%22%20parent%3D%221%22%3E%3CmxGeometry%20width%3D%2250%22%20height%3D%2250%22%20relative%3D%221%22%20as%3D%22geometry%22%3E%3CmxPoint%20x%3D%22160%22%20y%3D%22315%22%20as%3D%22sourcePoint%22%2F%3E%3CmxPoint%20x%3D%22230%22%20y%3D%22245%22%20as%3D%22targetPoint%22%2F%3E%3C%2FmxGeometry%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%2212%22%20value%3D%22%22%20style%3D%22endArrow%3Dclassic%3Bhtml%3D1%3Brounded%3D0%3BstrokeColor%3D%23FF3333%3BexitX%3D1%3BexitY%3D0.5%3BexitDx%3D0%3BexitDy%3D0%3BentryX%3D0%3BentryY%3D0.4%3BentryDx%3D0%3BentryDy%3D0%3BentryPerimeter%3D0%3B%22%20edge%3D%221%22%20source%3D%222%22%20target%3D%227%22%20parent%3D%221%22%3E%3CmxGeometry%20width%3D%2250%22%20height%3D%2250%22%20relative%3D%221%22%20as%3D%22geometry%22%3E%3CmxPoint%20x%3D%22170%22%20y%3D%22325%22%20as%3D%22sourcePoint%22%2F%3E%3CmxPoint%20x%3D%22240%22%20y%3D%22255%22%20as%3D%22targetPoint%22%2F%3E%3C%2FmxGeometry%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%2213%22%20value%3D%22%22%20style%3D%22endArrow%3Dclassic%3Bhtml%3D1%3Brounded%3D0%3BstrokeColor%3D%23FF3333%3BexitX%3D1%3BexitY%3D0.5%3BexitDx%3D0%3BexitDy%3D0%3BentryX%3D0%3BentryY%3D0.5%3BentryDx%3D0%3BentryDy%3D0%3B%22%20edge%3D%221%22%20source%3D%223%22%20target%3D%224%22%20parent%3D%221%22%3E%3CmxGeometry%20width%3D%2250%22%20height%3D%2250%22%20relative%3D%221%22%20as%3D%22geometry%22%3E%3CmxPoint%20x%3D%22180%22%20y%3D%22335%22%20as%3D%22sourcePoint%22%2F%3E%3CmxPoint%20x%3D%22250%22%20y%3D%22265%22%20as%3D%22targetPoint%22%2F%3E%3C%2FmxGeometry%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%2214%22%20value%3D%22%22%20style%3D%22endArrow%3Dclassic%3Bhtml%3D1%3Brounded%3D0%3BstrokeColor%3D%23FF3333%3BexitX%3D1%3BexitY%3D0.5%3BexitDx%3D0%3BexitDy%3D0%3BentryX%3D0%3BentryY%3D0.5%3BentryDx%3D0%3BentryDy%3D0%3B%22%20edge%3D%221%22%20source%3D%223%22%20target%3D%225%22%20parent%3D%221%22%3E%3CmxGeometry%20width%3D%2250%22%20height%3D%2250%22%20relative%3D%221%22%20as%3D%22geometry%22%3E%3CmxPoint%20x%3D%22190%22%20y%3D%22345%22%20as%3D%22sourcePoint%22%2F%3E%3CmxPoint%20x%3D%22260%22%20y%3D%22275%22%20as%3D%22targetPoint%22%2F%3E%3C%2FmxGeometry%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%2215%22%20value%3D%22%22%20style%3D%22endArrow%3Dclassic%3Bhtml%3D1%3Brounded%3D0%3BstrokeColor%3D%23FF3333%3BexitX%3D1%3BexitY%3D0.5%3BexitDx%3D0%3BexitDy%3D0%3BentryX%3D0%3BentryY%3D0.4%3BentryDx%3D0%3BentryDy%3D0%3BentryPerimeter%3D0%3B%22%20edge%3D%221%22%20source%3D%223%22%20target%3D%226%22%20parent%3D%221%22%3E%3CmxGeometry%20width%3D%2250%22%20height%3D%2250%22%20relative%3D%221%22%20as%3D%22geometry%22%3E%3CmxPoint%20x%3D%22200%22%20y%3D%22355%22%20as%3D%22sourcePoint%22%2F%3E%3CmxPoint%20x%3D%22270%22%20y%3D%22285%22%20as%3D%22targetPoint%22%2F%3E%3C%2FmxGeometry%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%2216%22%20value%3D%22%22%20style%3D%22endArrow%3Dclassic%3Bhtml%3D1%3Brounded%3D0%3BstrokeColor%3D%23FF3333%3BexitX%3D1%3BexitY%3D0.5%3BexitDx%3D0%3BexitDy%3D0%3B%22%20edge%3D%221%22%20source%3D%223%22%20parent%3D%221%22%3E%3CmxGeometry%20width%3D%2250%22%20height%3D%2250%22%20relative%3D%221%22%20as%3D%22geometry%22%3E%3CmxPoint%20x%3D%22210%22%20y%3D%22365%22%20as%3D%22sourcePoint%22%2F%3E%3CmxPoint%20x%3D%22210%22%20y%3D%22429.9999999999999%22%20as%3D%22targetPoint%22%2F%3E%3C%2FmxGeometry%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%2217%22%20value%3D%22%22%20style%3D%22endArrow%3Dclassic%3Bhtml%3D1%3Brounded%3D0%3BstrokeColor%3D%23FF3333%3B%22%20edge%3D%221%22%20source%3D%224%22%20parent%3D%221%22%3E%3CmxGeometry%20width%3D%2250%22%20height%3D%2250%22%20relative%3D%221%22%20as%3D%22geometry%22%3E%3CmxPoint%20x%3D%22300%22%20y%3D%22300%22%20as%3D%22sourcePoint%22%2F%3E%3CmxPoint%20x%3D%22320%22%20y%3D%22339.9999999999999%22%20as%3D%22targetPoint%22%2F%3E%3C%2FmxGeometry%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%2218%22%20value%3D%22%22%20style%3D%22endArrow%3Dclassic%3Bhtml%3D1%3Brounded%3D0%3BstrokeColor%3D%23FF3333%3BexitX%3D1.002%3BexitY%3D0.608%3BexitDx%3D0%3BexitDy%3D0%3BexitPerimeter%3D0%3BentryX%3D0.04%3BentryY%3D0.617%3BentryDx%3D0%3BentryDy%3D0%3BentryPerimeter%3D0%3B%22%20edge%3D%221%22%20source%3D%225%22%20target%3D%228%22%20parent%3D%221%22%3E%3CmxGeometry%20width%3D%2250%22%20height%3D%2250%22%20relative%3D%221%22%20as%3D%22geometry%22%3E%3CmxPoint%20x%3D%22230%22%20y%3D%22385%22%20as%3D%22sourcePoint%22%2F%3E%3CmxPoint%20x%3D%22280%22%20y%3D%22330%22%20as%3D%22targetPoint%22%2F%3E%3C%2FmxGeometry%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%2219%22%20value%3D%22%22%20style%3D%22endArrow%3Dclassic%3Bhtml%3D1%3Brounded%3D0%3BstrokeColor%3D%23FF3333%3BexitX%3D0.973%3BexitY%3D0.627%3BexitDx%3D0%3BexitDy%3D0%3BexitPerimeter%3D0%3BentryX%3D0%3BentryY%3D0.6%3BentryDx%3D0%3BentryDy%3D0%3BentryPerimeter%3D0%3B%22%20edge%3D%221%22%20source%3D%226%22%20target%3D%228%22%20parent%3D%221%22%3E%3CmxGeometry%20width%3D%2250%22%20height%3D%2250%22%20relative%3D%221%22%20as%3D%22geometry%22%3E%3CmxPoint%20x%3D%22240%22%20y%3D%22395%22%20as%3D%22sourcePoint%22%2F%3E%3CmxPoint%20x%3D%22290%22%20y%3D%22350%22%20as%3D%22targetPoint%22%2F%3E%3C%2FmxGeometry%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%2220%22%20value%3D%22%22%20style%3D%22endArrow%3Dclassic%3Bhtml%3D1%3Brounded%3D0%3BstrokeColor%3D%23FF3333%3BexitX%3D1%3BexitY%3D0.5%3BexitDx%3D0%3BexitDy%3D0%3BentryX%3D0%3BentryY%3D0.5%3BentryDx%3D0%3BentryDy%3D0%3B%22%20edge%3D%221%22%20source%3D%227%22%20target%3D%228%22%20parent%3D%221%22%3E%3CmxGeometry%20width%3D%2250%22%20height%3D%2250%22%20relative%3D%221%22%20as%3D%22geometry%22%3E%3CmxPoint%20x%3D%22290%22%20y%3D%22440%22%20as%3D%22sourcePoint%22%2F%3E%3CmxPoint%20x%3D%22330%22%20y%3D%22390%22%20as%3D%22targetPoint%22%2F%3E%3C%2FmxGeometry%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%2221%22%20value%3D%22Input%22%20style%3D%22text%3Bhtml%3D1%3Balign%3Dcenter%3BverticalAlign%3Dmiddle%3Bresizable%3D0%3Bpoints%3D%5B%5D%3Bautosize%3D1%3BstrokeColor%3Dnone%3BfillColor%3Dnone%3BfontFamily%3DVerdana%3BfontColor%3D%23CC0000%3BfontSize%3D14%3B%22%20vertex%3D%221%22%20parent%3D%221%22%3E%3CmxGeometry%20x%3D%2290%22%20y%3D%22240%22%20width%3D%2250%22%20height%3D%2220%22%20as%3D%22geometry%22%2F%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%2222%22%20value%3D%22Output%22%20style%3D%22text%3Bhtml%3D1%3Balign%3Dcenter%3BverticalAlign%3Dmiddle%3Bresizable%3D0%3Bpoints%3D%5B%5D%3Bautosize%3D1%3BstrokeColor%3Dnone%3BfillColor%3Dnone%3BfontFamily%3DVerdana%3BfontSize%3D14%3BfontColor%3D%23CC0000%3B%22%20vertex%3D%221%22%20parent%3D%221%22%3E%3CmxGeometry%20x%3D%22320%22%20y%3D%22280%22%20width%3D%2250%22%20height%3D%2220%22%20as%3D%22geometry%22%2F%3E%3C%2FmxCell%3E%3CmxCell%20id%3D%2223%22%20value%3D%22Intermediate%22%20style%3D%22text%3Bhtml%3D1%3Balign%3Dcenter%3BverticalAlign%3Dmiddle%3Bresizable%3D0%3Bpoints%3D%5B%5D%3Bautosize%3D1%3BstrokeColor%3Dnone%3BfillColor%3Dnone%3BfontFamily%3DVerdana%3BfontSize%3D14%3BfontColor%3D%23CC0000%3B%22%20vertex%3D%221%22%20parent%3D%221%22%3E%3CmxGeometry%20x%3D%22190%22%20y%3D%22170%22%20width%3D%2290%22%20height%3D%2220%22%20as%3D%22geometry%22%2F%3E%3C%2FmxCell%3E%3C%2Froot%3E%3C%2FmxGraphModelt

T

Anonymous
  • feiticeir0
    feiticeir0 12 days ago in reply to DAB

    After the quantization, no . We have the model ready for deployment in the microcontroller.

    Before the quantization, yes, we can tune those values - hyperparameters.

    • Cancel
    • Up 0 Down
    • Reply
    • More
    • Cancel
  • DAB
    DAB 13 days ago

    Is it possible to access the weights and biases??

    • Cancel
    • Up 0 Down
    • Reply
    • More
    • Cancel
Element14

element14 is the first online community specifically for engineers. Connect with your peers and get expert answers to your questions.

  • Members
  • Learn
  • Technologies
  • Challenges & Projects
  • Products
  • Store
  • About Us
  • Feedback & Support
  • FAQs
  • Terms of Use
  • Privacy Policy
  • Legal and Copyright Notices
  • Sitemap
  • Cookies

An Avnet Company © 2022 Premier Farnell Limited. All Rights Reserved.

Premier Farnell Ltd, registered in England and Wales (no 00876412), registered office: Farnell House, Forge Lane, Leeds LS12 2NE.

ICP 备案号 10220084.

Follow element14

  • Facebook
  • Twitter
  • linkedin
  • YouTube