Deep Learning with CNTK and F#

Overview

Explore deep learning techniques using Microsoft's Cognitive Toolkit (CNTK) and F# in this 57-minute conference talk. Gain insights into neural networks and their applications in computer vision, speech recognition, and automated language translation. Learn how to train and implement deep learning models within .NET environments. Discover the underlying concepts, including tensors, weights, and filters, while understanding the practical aspects of machine learning deployment. Follow along as the speaker demonstrates creating a system for rendering numbers on screen, implementing volume control, and working with datasets. Dive into code examples, explore the CNTK API, and witness a live demo showcasing model training and prediction. By the end of this talk, acquire the knowledge to leverage F# for effortless model learning and integration into your .NET projects.

Syllabus

Introduction
Who am I
Context
Plan
Deployment
Rendering Numbers on Screen
Creating a System
Volume Controller
Weights
Filters
Machine Learning
Deep Learning
CNTK
What is a tensor
How does this plugin work
Fully trained model
Variables
Code
Dataset
Model
Running the model
Results
API
Raw API
Models
Demo
Data Set
Small subset
change model
model file
model predictions
conclusion
thank you