Deep learning in web browsers with TensorFlow JS ,made easy, fast and deployable
A very latest era of Browser based machine learning has arrived. With the arrival of TensorFlow JS .This talk aims to introduce everyone with browser based Deep learning and how easy it is to embed and train models in browsers just using Javascript. With JS based ML its also pretty easy to now integrate machine learning models with your web applications. TensorFlow JS also enables us to create beautiful Deep learning Demos as well .This talk aims to provide a proper technical and functional introduction to tensorFlow JS,deep learning in it and also capabilities like transfer learning ,reinforcement learning, re training models in browser and also some cool demos of the same.
Outline/Structure of the Talk
- Browser based Deep Learning ,What is it?
- Introducing TensorFlow JS
- Endless Possibilities of TensorFlow JS
- Architecting ML Applications before TensorFlow JS
- Architecting Lite ML web applications with TensorFlow JS
- Demos using simple Javascript Deep learning
- Using Pre Trained models with Transfer learning
- How easy it is to code in TensorFlow JS
- Utilising GPUs and browser's powers using TF JS
- Python and JavaScript Deep learning comparison/connection
Learning Outcome
This talk is intended for every machine learning beginner ,practitioner who wants to develop and deploy lightweight machine learning models.Up until now ,people thought only R and python had the capability to do machine learning, with TensorFlow JS now possibilities are endless, easy machine learning, right in your browser without needing any heavy hardware
It also enables a developer to develop run and embed complex ML models right in the browser and also re train already existing models.
One also learn utilising browser's power via WebGL, node JS based training ,GPU usage and beautiful javascript visualisations in ML using JS
Target Audience
All Machine learning /Deep learning practitioners ,JavaScript developers ,Machine learning Developers
Prerequisites for Attendees
Basic Knowledge of Machine learning/Deep learning and Javascript
Video
Links
Umang Sharma Demoing Deep Learning based Pacman game built using TF js
My TensorFlow and cloud ML workshop quick video
Participant Testimonial about my Google's Machine learning Crash course Program
An open source TensorFlow JS plugin Developed by me for IntellIJ users
https://github.com/umang-sh/tensorFlowJS-Plugin
Invited Talks given :
Invited talks and lectures given so far:
"Machine learning and Paradigms of new IT" at Google's Women Techmakers Conference, Pune, March 2018.
"Maths and Theory behind machine learning" for Google's Machine learning crash course Pune, August 2018
"Machine learning with TensorFlow" for Google's Machine learning crash course Pune, August 2018
"Image Classification with Deep learning for Google' Machine learning crash course Pune, September 2018
"Neural networks with Theory and Implementation"for Google' Machine learning crash course Pune, September 2018
"Machine learning application development in Multiple languages " for Google' Machine learning crash course Pune, September 2018
"Why and How should machine learn from humans" at Google DevFest , Pune, November 2018
Lecture on "TensorFlow and Machine learning application development" at PyData Mumbai, September 2018
Keynote Talk at "Deep learning is your browsers made easy, fast and Deployable " at A.I fest College of Engineering Kerala ,March 2019.
TensorFlow JS : https://www.tensorflow.org/js/
Demos on TensorFlow JS :https://www.tensorflow.org/js/demos
My linkedIn profile:
https://www.linkedin.com/in/umang-sharma-321b4483/
My Website about me
schedule Submitted 3 years ago
People who liked this proposal, also liked:
-
keyboard_arrow_downAntrixsh Gupta - Creating Custom Interactive Data Visualization Dashboards with Bokeh
90 Mins
Workshop
Beginner
This will be a hands-on workshop how to build a custom interactive dashboard application on your local machine or on any cloud service provider. You will also learn how to deploy this application with both security and scalability in mind.
Powerful Data visualization software solutions are extremely useful when building interactive data visualization dashboards. However, these types of solutions might not provide sufficient customization options. For those scenarios, you can use open source libraries like D3.js, Chart.js, or Bokeh to create custom dashboards. While these libraries offer a lot of flexibility for building dashboards with tailored features and visualizations.
-
Recent Activities
- description Umang Sharma updated proposal Deep learning in web brow... 3 Timesschedule 3 years ago
- description Umang Sharma updated proposal Deep learning in browser... 6 Timesschedule 3 years ago
- description Umang Sharma updated proposal Deep learning in your bro...schedule 3 years ago
Call for Proposals CLOSED
Ended on Apr 15 '19 11:30 PM IST
We are looking for deep technical topics related to Data Science. We are committed to diversity and believe in transparency. Hence, all proposals will be public.
Registered user will be able to comment on your proposal. You are required to reply to those comments to provide clarifications, explain revisions and respond to questions. The program team will look at how well you've responded to comments and updated your proposal to incorporate public suggestions. Ultimately the decision to accept a session resides with the program team.
Your proposal stands the best chance to be selected, if it's unique, fully flushed, ready-to-go. Please ensure you've read:
To encourage early submissions and iterative improvement of proposals, we start accepting proposals as soon as we find them a good fit. Since competition gets tougher with progressing time, it is advisable to avoid waiting until the last minute for proposal submission.
Dr. Vikas Agrawal
Anoop Kulkarni
Naresh Jain
Usha Rengaraju
Dipanjan Sarkar
