SlideShare ist ein Scribd-Unternehmen logo
1 von 14
Supercharging your web apps with
ML using MediaPipe
Google I/O Extended, Warri
The challenge
As a web developer, I will like to apply ML to my web applications
The Solution?
Google MediaPipe!
What is MediaPipe
● Created by Google in 2012
● To be used for processing Youtube videos for compliance
● Eventually expanded to other Google products like Google Home
● First announced publicly in 2019
● It is a cross-platform, open source framework
● With a low-code API
MediaPipe solutions
Vision Solutions Text Solutions
More at https://developers.google.com/mediapipe/solutions
Using MediaPipe in your web apps
Demo app: https://gdg-warri.netlify.app/
Description: A movie rating website, which uses Sentiment Analysis a branch of
Text Classification to categorize text with predefined labels, and in this case
“positive” or “negative”.
Give your review about a movie to try it out!
Text Classification
Text classification is an NLP task where algorithms automatically assign
predefined categories or labels to text documents based on their content.
Creating a Text Classifier in 3 steps
1. Install the @mediapipe/tasks-text package
2. Load WebAssembly files needed for the classification
3. Instantiate the TextClassifier with custom options
Step one:
Install the MediaPipe library just like any other JS library..
Step two:
Import the Web Assembly files MediaPipe needs to work on Text Tasks
modelAssetPath: Path to the model our classification is going to run on.
maxResults: The number of top results to return.
scoreThreshold: Percentage at which results should be returned
Text Classification models
● BERT-classifier model (recommended for performance)
● Average word embedding model
● …and yours!
Learn more: https://developers.google.com/mediapipe/solutions/text/text_classifier#models
About me
Connect with me
Samuel Olaegbe
Twitter: @devloader
Github: @goodhands
LinkedIn: Samuel Olaegbe
Telegram: https://t.me/samuelolaegbe
Blog: https://devloader.hashnode.dev/

Weitere ähnliche Inhalte

Ähnlich wie Supercharging your web apps with ML using MediaPipe.pptx

Evaluation question (final1): 4
Evaluation question (final1): 4Evaluation question (final1): 4
Evaluation question (final1): 4jonathan randall
 
Jetpack compose session1 (1).pptx
Jetpack compose session1 (1).pptxJetpack compose session1 (1).pptx
Jetpack compose session1 (1).pptxShubhamJogdand8
 
Learn Django Tips, Tricks & Techniques for Developers
Learn Django Tips, Tricks & Techniques for DevelopersLearn Django Tips, Tricks & Techniques for Developers
Learn Django Tips, Tricks & Techniques for DevelopersMars Devs
 
Introduction to Google Cloud Endpoints: Speed Up Your API Development
Introduction to Google Cloud Endpoints: Speed Up Your API DevelopmentIntroduction to Google Cloud Endpoints: Speed Up Your API Development
Introduction to Google Cloud Endpoints: Speed Up Your API DevelopmentColin Su
 
How did you use media technologies in theconstruction, research, planning and...
How did you use media technologies in theconstruction, research, planning and...How did you use media technologies in theconstruction, research, planning and...
How did you use media technologies in theconstruction, research, planning and...vishal-s93
 
Gaining the app visibility that matters
Gaining the app visibility that mattersGaining the app visibility that matters
Gaining the app visibility that mattersPrajyot Mainkar
 
InMobi inDecode - Gaining App Visibility That Matters
InMobi inDecode - Gaining App Visibility That MattersInMobi inDecode - Gaining App Visibility That Matters
InMobi inDecode - Gaining App Visibility That MattersInMobi
 
Evaluation: Question Six
Evaluation: Question SixEvaluation: Question Six
Evaluation: Question Sixmillerjess
 
Getting SEO & Social Media Alignment Inside Your Organization by Warren Lee
Getting SEO & Social Media Alignment Inside Your Organization by Warren LeeGetting SEO & Social Media Alignment Inside Your Organization by Warren Lee
Getting SEO & Social Media Alignment Inside Your Organization by Warren LeeSearch Marketing Expo - SMX
 
How to Access and Use Gemini API for Free.pptx
How to Access and Use Gemini API for Free.pptxHow to Access and Use Gemini API for Free.pptx
How to Access and Use Gemini API for Free.pptxujjwalsoni23
 
Insights on Android App Development, Marketing and Monetization
Insights on Android App Development, Marketing and MonetizationInsights on Android App Development, Marketing and Monetization
Insights on Android App Development, Marketing and MonetizationJayneel Patel
 
Perspective presentation
Perspective presentationPerspective presentation
Perspective presentationOskar Hargedahl
 
Evaluation question: 4 (Updated)
Evaluation question: 4 (Updated)Evaluation question: 4 (Updated)
Evaluation question: 4 (Updated)jonathan randall
 
ICT I intend on using to create my trailer.
ICT I intend on using to create my trailer.ICT I intend on using to create my trailer.
ICT I intend on using to create my trailer.AjayFlyguy
 

Ähnlich wie Supercharging your web apps with ML using MediaPipe.pptx (20)

Evaluation question (final1): 4
Evaluation question (final1): 4Evaluation question (final1): 4
Evaluation question (final1): 4
 
Evaluation Question 4
Evaluation Question 4Evaluation Question 4
Evaluation Question 4
 
Evaluation 4
Evaluation 4Evaluation 4
Evaluation 4
 
Jetpack compose session1 (1).pptx
Jetpack compose session1 (1).pptxJetpack compose session1 (1).pptx
Jetpack compose session1 (1).pptx
 
Learn Django Tips, Tricks & Techniques for Developers
Learn Django Tips, Tricks & Techniques for DevelopersLearn Django Tips, Tricks & Techniques for Developers
Learn Django Tips, Tricks & Techniques for Developers
 
Evaluation Question: 4
Evaluation Question: 4Evaluation Question: 4
Evaluation Question: 4
 
Google IO 2018
Google IO 2018Google IO 2018
Google IO 2018
 
Introduction to Google Cloud Endpoints: Speed Up Your API Development
Introduction to Google Cloud Endpoints: Speed Up Your API DevelopmentIntroduction to Google Cloud Endpoints: Speed Up Your API Development
Introduction to Google Cloud Endpoints: Speed Up Your API Development
 
Evaluation 4
Evaluation 4Evaluation 4
Evaluation 4
 
How did you use media technologies in theconstruction, research, planning and...
How did you use media technologies in theconstruction, research, planning and...How did you use media technologies in theconstruction, research, planning and...
How did you use media technologies in theconstruction, research, planning and...
 
Gaining the app visibility that matters
Gaining the app visibility that mattersGaining the app visibility that matters
Gaining the app visibility that matters
 
InMobi inDecode - Gaining App Visibility That Matters
InMobi inDecode - Gaining App Visibility That MattersInMobi inDecode - Gaining App Visibility That Matters
InMobi inDecode - Gaining App Visibility That Matters
 
Evaluation: Question Six
Evaluation: Question SixEvaluation: Question Six
Evaluation: Question Six
 
Getting SEO & Social Media Alignment Inside Your Organization by Warren Lee
Getting SEO & Social Media Alignment Inside Your Organization by Warren LeeGetting SEO & Social Media Alignment Inside Your Organization by Warren Lee
Getting SEO & Social Media Alignment Inside Your Organization by Warren Lee
 
How to Access and Use Gemini API for Free.pptx
How to Access and Use Gemini API for Free.pptxHow to Access and Use Gemini API for Free.pptx
How to Access and Use Gemini API for Free.pptx
 
Insights on Android App Development, Marketing and Monetization
Insights on Android App Development, Marketing and MonetizationInsights on Android App Development, Marketing and Monetization
Insights on Android App Development, Marketing and Monetization
 
Evaluation question; 4
Evaluation question; 4Evaluation question; 4
Evaluation question; 4
 
Perspective presentation
Perspective presentationPerspective presentation
Perspective presentation
 
Evaluation question: 4 (Updated)
Evaluation question: 4 (Updated)Evaluation question: 4 (Updated)
Evaluation question: 4 (Updated)
 
ICT I intend on using to create my trailer.
ICT I intend on using to create my trailer.ICT I intend on using to create my trailer.
ICT I intend on using to create my trailer.
 

Mehr von Samuel Olaegbe

ENABLING INCLUSIVE FINTECH IN THE NIGERIAN CAPITAL MARKET: A REGULATORS PERSP...
ENABLING INCLUSIVE FINTECH IN THE NIGERIAN CAPITAL MARKET: A REGULATORS PERSP...ENABLING INCLUSIVE FINTECH IN THE NIGERIAN CAPITAL MARKET: A REGULATORS PERSP...
ENABLING INCLUSIVE FINTECH IN THE NIGERIAN CAPITAL MARKET: A REGULATORS PERSP...Samuel Olaegbe
 
Regulatory framework for lending: State v. Federal perspectives
Regulatory framework for lending: State v. Federal perspectivesRegulatory framework for lending: State v. Federal perspectives
Regulatory framework for lending: State v. Federal perspectivesSamuel Olaegbe
 
How data analytics will drive the future of banking
How data analytics will drive the future of bankingHow data analytics will drive the future of banking
How data analytics will drive the future of bankingSamuel Olaegbe
 
BROADBAND PENETRATION IN NIGERIA
BROADBAND PENETRATION IN NIGERIABROADBAND PENETRATION IN NIGERIA
BROADBAND PENETRATION IN NIGERIASamuel Olaegbe
 
How Should Fintech be regulated?
How Should Fintech be regulated?How Should Fintech be regulated?
How Should Fintech be regulated?Samuel Olaegbe
 
Payment Services Bank; the challenger banks
Payment Services Bank; the challenger banksPayment Services Bank; the challenger banks
Payment Services Bank; the challenger banksSamuel Olaegbe
 
Embracing Open Banking as an Opportunity to Innovate, not just an Exercise in...
Embracing Open Banking as an Opportunity to Innovate, not just an Exercise in...Embracing Open Banking as an Opportunity to Innovate, not just an Exercise in...
Embracing Open Banking as an Opportunity to Innovate, not just an Exercise in...Samuel Olaegbe
 
Building infrastructure for_collaborative_fintech_3.0
Building infrastructure for_collaborative_fintech_3.0Building infrastructure for_collaborative_fintech_3.0
Building infrastructure for_collaborative_fintech_3.0Samuel Olaegbe
 

Mehr von Samuel Olaegbe (8)

ENABLING INCLUSIVE FINTECH IN THE NIGERIAN CAPITAL MARKET: A REGULATORS PERSP...
ENABLING INCLUSIVE FINTECH IN THE NIGERIAN CAPITAL MARKET: A REGULATORS PERSP...ENABLING INCLUSIVE FINTECH IN THE NIGERIAN CAPITAL MARKET: A REGULATORS PERSP...
ENABLING INCLUSIVE FINTECH IN THE NIGERIAN CAPITAL MARKET: A REGULATORS PERSP...
 
Regulatory framework for lending: State v. Federal perspectives
Regulatory framework for lending: State v. Federal perspectivesRegulatory framework for lending: State v. Federal perspectives
Regulatory framework for lending: State v. Federal perspectives
 
How data analytics will drive the future of banking
How data analytics will drive the future of bankingHow data analytics will drive the future of banking
How data analytics will drive the future of banking
 
BROADBAND PENETRATION IN NIGERIA
BROADBAND PENETRATION IN NIGERIABROADBAND PENETRATION IN NIGERIA
BROADBAND PENETRATION IN NIGERIA
 
How Should Fintech be regulated?
How Should Fintech be regulated?How Should Fintech be regulated?
How Should Fintech be regulated?
 
Payment Services Bank; the challenger banks
Payment Services Bank; the challenger banksPayment Services Bank; the challenger banks
Payment Services Bank; the challenger banks
 
Embracing Open Banking as an Opportunity to Innovate, not just an Exercise in...
Embracing Open Banking as an Opportunity to Innovate, not just an Exercise in...Embracing Open Banking as an Opportunity to Innovate, not just an Exercise in...
Embracing Open Banking as an Opportunity to Innovate, not just an Exercise in...
 
Building infrastructure for_collaborative_fintech_3.0
Building infrastructure for_collaborative_fintech_3.0Building infrastructure for_collaborative_fintech_3.0
Building infrastructure for_collaborative_fintech_3.0
 

Kürzlich hochgeladen

Effective Strategies for Wix's Scaling challenges - GeeCon
Effective Strategies for Wix's Scaling challenges - GeeConEffective Strategies for Wix's Scaling challenges - GeeCon
Effective Strategies for Wix's Scaling challenges - GeeConNatan Silnitsky
 
The mythical technical debt. (Brooke, please, forgive me)
The mythical technical debt. (Brooke, please, forgive me)The mythical technical debt. (Brooke, please, forgive me)
The mythical technical debt. (Brooke, please, forgive me)Roberto Bettazzoni
 
Novo Nordisk: When Knowledge Graphs meet LLMs
Novo Nordisk: When Knowledge Graphs meet LLMsNovo Nordisk: When Knowledge Graphs meet LLMs
Novo Nordisk: When Knowledge Graphs meet LLMsNeo4j
 
Artyushina_Guest lecture_YorkU CS May 2024.pptx
Artyushina_Guest lecture_YorkU CS May 2024.pptxArtyushina_Guest lecture_YorkU CS May 2024.pptx
Artyushina_Guest lecture_YorkU CS May 2024.pptxAnnaArtyushina1
 
Lessons Learned from Building a Serverless Notifications System.pdf
Lessons Learned from Building a Serverless Notifications System.pdfLessons Learned from Building a Serverless Notifications System.pdf
Lessons Learned from Building a Serverless Notifications System.pdfSrushith Repakula
 
Transformer Neural Network Use Cases with Links
Transformer Neural Network Use Cases with LinksTransformer Neural Network Use Cases with Links
Transformer Neural Network Use Cases with LinksJinanKordab
 
GraphSummit Milan - Visione e roadmap del prodotto Neo4j
GraphSummit Milan - Visione e roadmap del prodotto Neo4jGraphSummit Milan - Visione e roadmap del prodotto Neo4j
GraphSummit Milan - Visione e roadmap del prodotto Neo4jNeo4j
 
Test Automation Design Patterns_ A Comprehensive Guide.pdf
Test Automation Design Patterns_ A Comprehensive Guide.pdfTest Automation Design Patterns_ A Comprehensive Guide.pdf
Test Automation Design Patterns_ A Comprehensive Guide.pdfkalichargn70th171
 
Workshop - Architecting Innovative Graph Applications- GraphSummit Milan
Workshop -  Architecting Innovative Graph Applications- GraphSummit MilanWorkshop -  Architecting Innovative Graph Applications- GraphSummit Milan
Workshop - Architecting Innovative Graph Applications- GraphSummit MilanNeo4j
 
CERVED e Neo4j su una nuvola, migrazione ed evoluzione di un grafo mission cr...
CERVED e Neo4j su una nuvola, migrazione ed evoluzione di un grafo mission cr...CERVED e Neo4j su una nuvola, migrazione ed evoluzione di un grafo mission cr...
CERVED e Neo4j su una nuvola, migrazione ed evoluzione di un grafo mission cr...Neo4j
 
Evolving Data Governance for the Real-time Streaming and AI Era
Evolving Data Governance for the Real-time Streaming and AI EraEvolving Data Governance for the Real-time Streaming and AI Era
Evolving Data Governance for the Real-time Streaming and AI Eraconfluent
 
GraphSummit Milan - Neo4j: The Art of the Possible with Graph
GraphSummit Milan - Neo4j: The Art of the Possible with GraphGraphSummit Milan - Neo4j: The Art of the Possible with Graph
GraphSummit Milan - Neo4j: The Art of the Possible with GraphNeo4j
 
Wired_2.0_CREATE YOUR ULTIMATE LEARNING ENVIRONMENT_JCON_16052024
Wired_2.0_CREATE YOUR ULTIMATE LEARNING ENVIRONMENT_JCON_16052024Wired_2.0_CREATE YOUR ULTIMATE LEARNING ENVIRONMENT_JCON_16052024
Wired_2.0_CREATE YOUR ULTIMATE LEARNING ENVIRONMENT_JCON_16052024SimonedeGijt
 
From Theory to Practice: Utilizing SpiraPlan's REST API
From Theory to Practice: Utilizing SpiraPlan's REST APIFrom Theory to Practice: Utilizing SpiraPlan's REST API
From Theory to Practice: Utilizing SpiraPlan's REST APIInflectra
 
BusinessGPT - Security and Governance for Generative AI
BusinessGPT  - Security and Governance for Generative AIBusinessGPT  - Security and Governance for Generative AI
BusinessGPT - Security and Governance for Generative AIAGATSoftware
 

Kürzlich hochgeladen (20)

Effective Strategies for Wix's Scaling challenges - GeeCon
Effective Strategies for Wix's Scaling challenges - GeeConEffective Strategies for Wix's Scaling challenges - GeeCon
Effective Strategies for Wix's Scaling challenges - GeeCon
 
The mythical technical debt. (Brooke, please, forgive me)
The mythical technical debt. (Brooke, please, forgive me)The mythical technical debt. (Brooke, please, forgive me)
The mythical technical debt. (Brooke, please, forgive me)
 
Novo Nordisk: When Knowledge Graphs meet LLMs
Novo Nordisk: When Knowledge Graphs meet LLMsNovo Nordisk: When Knowledge Graphs meet LLMs
Novo Nordisk: When Knowledge Graphs meet LLMs
 
Artyushina_Guest lecture_YorkU CS May 2024.pptx
Artyushina_Guest lecture_YorkU CS May 2024.pptxArtyushina_Guest lecture_YorkU CS May 2024.pptx
Artyushina_Guest lecture_YorkU CS May 2024.pptx
 
Abortion Pill Prices Aliwal North ](+27832195400*)[ 🏥 Women's Abortion Clinic...
Abortion Pill Prices Aliwal North ](+27832195400*)[ 🏥 Women's Abortion Clinic...Abortion Pill Prices Aliwal North ](+27832195400*)[ 🏥 Women's Abortion Clinic...
Abortion Pill Prices Aliwal North ](+27832195400*)[ 🏥 Women's Abortion Clinic...
 
微信号购买
微信号购买微信号购买
微信号购买
 
Lessons Learned from Building a Serverless Notifications System.pdf
Lessons Learned from Building a Serverless Notifications System.pdfLessons Learned from Building a Serverless Notifications System.pdf
Lessons Learned from Building a Serverless Notifications System.pdf
 
Abortion Clinic In Pretoria ](+27832195400*)[ 🏥 Safe Abortion Pills in Pretor...
Abortion Clinic In Pretoria ](+27832195400*)[ 🏥 Safe Abortion Pills in Pretor...Abortion Clinic In Pretoria ](+27832195400*)[ 🏥 Safe Abortion Pills in Pretor...
Abortion Clinic In Pretoria ](+27832195400*)[ 🏥 Safe Abortion Pills in Pretor...
 
Transformer Neural Network Use Cases with Links
Transformer Neural Network Use Cases with LinksTransformer Neural Network Use Cases with Links
Transformer Neural Network Use Cases with Links
 
GraphSummit Milan - Visione e roadmap del prodotto Neo4j
GraphSummit Milan - Visione e roadmap del prodotto Neo4jGraphSummit Milan - Visione e roadmap del prodotto Neo4j
GraphSummit Milan - Visione e roadmap del prodotto Neo4j
 
Abortion Clinic In Johannesburg ](+27832195400*)[ 🏥 Safe Abortion Pills in Jo...
Abortion Clinic In Johannesburg ](+27832195400*)[ 🏥 Safe Abortion Pills in Jo...Abortion Clinic In Johannesburg ](+27832195400*)[ 🏥 Safe Abortion Pills in Jo...
Abortion Clinic In Johannesburg ](+27832195400*)[ 🏥 Safe Abortion Pills in Jo...
 
Test Automation Design Patterns_ A Comprehensive Guide.pdf
Test Automation Design Patterns_ A Comprehensive Guide.pdfTest Automation Design Patterns_ A Comprehensive Guide.pdf
Test Automation Design Patterns_ A Comprehensive Guide.pdf
 
Workshop - Architecting Innovative Graph Applications- GraphSummit Milan
Workshop -  Architecting Innovative Graph Applications- GraphSummit MilanWorkshop -  Architecting Innovative Graph Applications- GraphSummit Milan
Workshop - Architecting Innovative Graph Applications- GraphSummit Milan
 
Abortion Pill Prices Germiston ](+27832195400*)[ 🏥 Women's Abortion Clinic in...
Abortion Pill Prices Germiston ](+27832195400*)[ 🏥 Women's Abortion Clinic in...Abortion Pill Prices Germiston ](+27832195400*)[ 🏥 Women's Abortion Clinic in...
Abortion Pill Prices Germiston ](+27832195400*)[ 🏥 Women's Abortion Clinic in...
 
CERVED e Neo4j su una nuvola, migrazione ed evoluzione di un grafo mission cr...
CERVED e Neo4j su una nuvola, migrazione ed evoluzione di un grafo mission cr...CERVED e Neo4j su una nuvola, migrazione ed evoluzione di un grafo mission cr...
CERVED e Neo4j su una nuvola, migrazione ed evoluzione di un grafo mission cr...
 
Evolving Data Governance for the Real-time Streaming and AI Era
Evolving Data Governance for the Real-time Streaming and AI EraEvolving Data Governance for the Real-time Streaming and AI Era
Evolving Data Governance for the Real-time Streaming and AI Era
 
GraphSummit Milan - Neo4j: The Art of the Possible with Graph
GraphSummit Milan - Neo4j: The Art of the Possible with GraphGraphSummit Milan - Neo4j: The Art of the Possible with Graph
GraphSummit Milan - Neo4j: The Art of the Possible with Graph
 
Wired_2.0_CREATE YOUR ULTIMATE LEARNING ENVIRONMENT_JCON_16052024
Wired_2.0_CREATE YOUR ULTIMATE LEARNING ENVIRONMENT_JCON_16052024Wired_2.0_CREATE YOUR ULTIMATE LEARNING ENVIRONMENT_JCON_16052024
Wired_2.0_CREATE YOUR ULTIMATE LEARNING ENVIRONMENT_JCON_16052024
 
From Theory to Practice: Utilizing SpiraPlan's REST API
From Theory to Practice: Utilizing SpiraPlan's REST APIFrom Theory to Practice: Utilizing SpiraPlan's REST API
From Theory to Practice: Utilizing SpiraPlan's REST API
 
BusinessGPT - Security and Governance for Generative AI
BusinessGPT  - Security and Governance for Generative AIBusinessGPT  - Security and Governance for Generative AI
BusinessGPT - Security and Governance for Generative AI
 

Supercharging your web apps with ML using MediaPipe.pptx

  • 1. Supercharging your web apps with ML using MediaPipe Google I/O Extended, Warri
  • 2. The challenge As a web developer, I will like to apply ML to my web applications
  • 4. What is MediaPipe ● Created by Google in 2012 ● To be used for processing Youtube videos for compliance ● Eventually expanded to other Google products like Google Home ● First announced publicly in 2019 ● It is a cross-platform, open source framework ● With a low-code API
  • 5. MediaPipe solutions Vision Solutions Text Solutions More at https://developers.google.com/mediapipe/solutions
  • 6. Using MediaPipe in your web apps Demo app: https://gdg-warri.netlify.app/ Description: A movie rating website, which uses Sentiment Analysis a branch of Text Classification to categorize text with predefined labels, and in this case “positive” or “negative”. Give your review about a movie to try it out!
  • 7. Text Classification Text classification is an NLP task where algorithms automatically assign predefined categories or labels to text documents based on their content.
  • 8. Creating a Text Classifier in 3 steps 1. Install the @mediapipe/tasks-text package 2. Load WebAssembly files needed for the classification 3. Instantiate the TextClassifier with custom options
  • 9. Step one: Install the MediaPipe library just like any other JS library..
  • 10. Step two: Import the Web Assembly files MediaPipe needs to work on Text Tasks
  • 11. modelAssetPath: Path to the model our classification is going to run on. maxResults: The number of top results to return. scoreThreshold: Percentage at which results should be returned
  • 12. Text Classification models ● BERT-classifier model (recommended for performance) ● Average word embedding model ● …and yours! Learn more: https://developers.google.com/mediapipe/solutions/text/text_classifier#models
  • 14. Connect with me Samuel Olaegbe Twitter: @devloader Github: @goodhands LinkedIn: Samuel Olaegbe Telegram: https://t.me/samuelolaegbe Blog: https://devloader.hashnode.dev/

Hinweis der Redaktion

  1. Facial detection Object identification Gesture detection Facial posture Face Landmark Detection Text classification Language detection
  2. You could use the remote URL to the model instead: https://storage.googleapis.com/mediapipe-models/text_classifier/bert_classifier/float32/latest/bert_classifier.tflite
  3. BERT-classifier model; This model uses a BERT-based architecture (specifically, the MobileBERT model) and is recommended because of its high accuracy. It contains metadata that allows the task to perform out-of-graph BERT tokenization. - Bidirectional Encoder Representations from Transformers Average word embedding model: This model uses an average word-embedding architecture. This model offers a smaller model size and lower latency at the cost of a lower prediction accuracy compared to the BERT-classifier. Customizing this model through additional training is also faster than doing training of the BERT-based classifier. This model contains metadata that allows the task to perform out-of-graph regex tokenization.