Unleash Deep Learning: Begin Visually with Caffe and DIGITS
4.5 (42 ratings)
Instead of using a simple lifetime average, Udemy calculates a course's star rating by considering a number of different factors such as the number of ratings, the age of ratings, and the likelihood of fraudulent ratings.
2,902 students enrolled
Wishlisted Wishlist

Please confirm that you want to add Unleash Deep Learning: Begin Visually with Caffe and DIGITS to your Wishlist.

Add to Wishlist

Unleash Deep Learning: Begin Visually with Caffe and DIGITS

An introduction to Deep Learning tools using Caffe and DIGITS where you get to create your own Deep Learning Model
4.5 (42 ratings)
Instead of using a simple lifetime average, Udemy calculates a course's star rating by considering a number of different factors such as the number of ratings, the age of ratings, and the likelihood of fraudulent ratings.
2,902 students enrolled
Created by Razvan Pistolea
Last updated 9/2016
English
Current price: $10 Original price: $150 Discount: 93% off
5 hours left at this price!
30-Day Money-Back Guarantee
Includes:
  • 1 hour on-demand video
  • 1 Article
  • Full lifetime access
  • Access on mobile and TV
  • Certificate of Completion
What Will I Learn?
  • Begin Deep Learning with a visual training method to increase ease and understanding
  • Create an LMDB database from the MNIST handwritten digits
  • Design a DNN with different layers like convolutions and pooling (subsampling)
  • Train a Deep Neural Network on CUDA enabled GPU
  • Deploy and Classify unseen test images
  • Evaluate the performance (top 1 and top 5) accuracy and confusion matrix of your model
View Curriculum
Requirements
  • Ubuntu 14.04 or higher
  • install DIGITS deep learning framework (Caffe is included)
Description

Learn the basics of Deep Learning with hands on exercises using the Caffe deep learning framework and the DIGITS visual interface. Build your own model and start classifying images.

Begin with a visual understanding of machine learning and deep learning concepts with this quick dive tutorial for beginners.

  • image classification
  • feedfoward neural network
  • convolutional neural network
  • digit recognition


A hot new topic with lots of opportunities

Artificial intelligence, machine learning and deep learning are in the news and all around us.  They give us the promise of computers solving tasks that until recently were very hard for computers: speech recognition, translation, object recognition, image classification, autonomous driving cars. 

Caffe framework is free, open sourced, continuously improved, has good documentation and even has an entire zoo of pre trained deep neural network models for image classification and other computer vision tasks. It is very fast and extensible and has most layers and utilities one could hope for (convolutions, pooling, relu, softmax, accuracy) so all you have to do is understand how to control this powerful tool.

DIGITS is NVIDIA's tool to help improve the process of designing, debugging and visualizing the inner workings of a deep neural network and works perfectly with Caffe.

The underling idea is very simple: instead of explicitly programming one should give lots of labeled examples and allow the computer to learn.


Content and Overview 

Suitable for beginning deep learning engineers. 

Thanks to DIGITS and Caffe there is a little programming and a lot of visual steps but a good mathematical and programming background is recommended.

You should already have Ubuntu (recommended) and DIGITS installed (a fork of Caffe will be included with the DIGITS installation).

The course will take you through the natural steps of getting training and testing data, designing a model, training the model and evaluating it.

Students completing the course will have the knowledge and courage to experiment and create amazing, useful and functional Convolutional Deep Learning Networks.


Who is the target audience?
  • SHOULD: students that want to begin Deep Learning with concepts and tools
  • SHOULD: students who want to learn and gain insights into why Deep Neural Networks are such a powerful and unique tool
  • SHOULD NOT: experts in Deep Learning
Students Who Viewed This Course Also Viewed
Curriculum For This Course
13 Lectures
01:03:31
+
Introduction and Overview
1 Lecture 01:39

toy deep learning network

generalize to bigger real life problems

graphical interface called DIIGITS


Preview 01:39

Basic Machine Learning
2 questions
+
App 1: Your first Deep Neural Network
7 Lectures 35:18

echo DIGITS_HOME

download mnist, uncompress, read labels


Overview and Download MNIST
04:17

Deep Learning GPU Training System (DIGITS) by NVIDIA

Lightning Memory-Mapped Database (LMDB) 

supervized vs unsupervized learning

training vs validation vs testing


web server, jobs

25% for validation

over fitting

class imbalance


Preview 06:35

convolution

pooling, sub sampling

full connection, inner product

automatic check reader

generalize the lesson to real problem

epoch

gradient of the error, stochastic gradient descent

base learning rate

softmax

accuracy


Preview 05:39

job directory

watch nvidia-smi

loss graph

accuracy graph


Step 4: Train Lenet
03:10

classify unseen handwritten digits

single, multiple images

confusion matrix

top-1 accuracy

top-5 accuracy

top N predictions per category


Step 5: Classify using the trained Lenet model
06:09

channels, height, width

convolution

matrix

pooling

feature maps


Preview 08:42

Summary
00:46
+
App 2: A simple custom DNN
3 Lectures 18:31

ipython notebook

github

fully connected feedforward neural network

predict

bottom top

number of outputs

confusion matrix



Customize Lenet based on the simple ANN course
09:23

solver mode: GPU

solver type: SGC

train

mean file

deploy

softmax

caffe log output

iteration number

variable learn rate

Preview 07:16

Summary
01:52
+
Bonus
2 Lectures 08:03

DIGITS archive

get_net

channels == 1

forward_pass

Deploy a trained model without using the DIGITS interface
07:56

Machine Learning Series

Part 1: Unleash Machine Learning: Build Artificial Neuron in Python

Part 2: Unleash Deep Learning: Begin Visually using Caffe and DIGITS (this course)

Part 3: Coming soon

Preview 00:07
About the Instructor
Razvan Pistolea
4.1 Average rating
250 Reviews
9,445 Students
4 Courses
Source Code Painter

I am a Machine Learning Engineer, Deep Learning Engineer and even an Indie Game Developer with a Major in Compilers and a Master's degree in Artificial Intelligence from University Politehnica of Bucharest.

I am passionate about Games and Artificial Intelligence. I love to give life to A.I. agents in my project or my friend's projects and I want to teach you too.