2. 2
DEEP LEARNING WITH CUDNN
cuDNN is a library for deep learning primitives
GPUs
cuDNN
Frameworks
Applications
Tesla TX-1 Titan
CPU is 16 core Haswell E5-2698 at 2.3 GHz, with 3.6
GHz Turbo
1 1
9
8
17
16
AlexNet GoogleNet
Speedup
CPU
Caffe GPU
Caffe+cuDNN
Titan X GPU
3. 3
CUDNN V3 RC AVAILABLE TODAY
Library for DNN toolkit developer and researchers
Contains building blocks for DNN toolkits
Convolutions, pooling, activation functions, etc.
Great performance and easy to deploy
developer.nvidia.com/cuDNN
cuBLAS (SGEMM for fully-connected layers) is part of CUDA toolkit,
developer.nvidia.com/cuda-toolkit
cuDNN (and cuBLAS)
4. 4
Process Data Configure DNN VisualizationMonitor Progress
Interactive Deep Learning GPU Training System
NVIDIA DIGITS
5. 5
NVIDIA DIGITS
Data Scientists & Researchers:
Quickly design the best deep neural
network (DNN) for your data
Visually monitor DNN training quality
in real-time
Manage training of many DNNs in
parallel on multi-GPU systems
Open source!
https://developer.nvidia.com/digits
Interactive Deep Learning GPU Training System
7. 7
Image parameter
options
Create your
dataset
DIGITS can automatically create your
training and validation set
NVIDIA DIGITS
Create Database
/path/to/images/
|-cat/
| |-cat1.jpg
| |-cat2.jpg
|-dog/
| |dog1.jpg
| |dog2.jpg
|-lizard/
|-lizard1.jpg
|- lizard2.jpg
Input Data Format
11. 11
NVIDIA DIGITS
Training Speedup Achieved with DIGITS on Multiple GeForce TITAN X GPUs in a DIGITS
DevBox. These results were obtained with the Caffe framework and a batch size of 128.
Possible speed up with
multiple GPUs
20. 20
NVIDIA DIGITS
Where to get DIGITS 2
Easy to use web installer https://developer.nvidia.com/digits
github - https://github.com/NVIDIA/DIGITS
Remember to install NVIDIA’s Caffe branch - https://github.com/NVIDIA/caffe
User support
DIGITS Users Google group - https://groups.google.com/forum/#!forum/digits-users
For more information on getting started with DIGITS
Parallel forall blogs - http://devblogs.nvidia.com/parallelforall/easy-multi-gpu-deep-learning-
digits-2/
Getting started guide - https://github.com/NVIDIA/DIGITS/blob/master/docs/GettingStarted.md
Resources
21. 21
INTERESTED IN LEARNING MORE ABOUT DEEP
LEARNING?
Check out our Free Deep Learning Courses -
https://developer.nvidia.com/deep-learning-courses
Date Class
7/22 Class #1 - Introduction to Deep Learning
8/5 Class #2 - Getting Started with DIGITS interactive training system for image
classification
8/19 Class #3 - Getting Started with the Caffe Framework
9/2 Class #4 - Getting Started with the Theano Framework
9/16 Class #5 - Getting Started with the Torch Framework
Office hours for Q&A too!