- TensorFlow is Google's open source machine learning library for developing and training neural networks and deep learning models. It operates using data flow graphs to represent computation.
- TensorFlow can be used across many platforms including data centers, CPUs, GPUs, mobile phones, and IoT devices. It is widely used at Google across many products and research areas involving machine learning.
- The TensorFlow library is used along with higher level tools in Google's machine learning platform including TensorFlow Cloud, Machine Learning APIs, and Cloud Machine Learning Platform to make machine learning more accessible and scalable.
6. Human:
Someone is using a small
grill to melt his sandwich.
Neural Model:
A person is cooking some
food on a grill.
7. One example (of 100s): “Neural Art” in TensorFlow
An implementation of
"A neural algorithm of Artistic
style" in TensorFlow, for
● Introductory, hackable demos
for TensorFlow, and
● Demonstrating the use of
importing various Caffe
cnn models (VGG and
illustration2vec) in TF.
github.com/woodrush/neural-art-tf
8. Growing Use of Deep Learning at Google
Number of directories containing model description files
UniqueProjectDirectories
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3
2000
1500
1000
500
0
Across many products/areas
● Apps
● Maps
● Photos
● Gmail
● Speech
● Android
● YouTube
● Translation
● Robotics Research
● Image Understanding
● Natural Language
Understanding
● Drug Discovery
2012 2013 2014 2015
Q4
9. The Machine Learning Spectrum
TensorFlow Cloud Machine Learning Machine Learning APIs
Data Scientist
15. Dataflow Graph (backward graph and updates)
Backward graph and update are added automatically to graph
Add Mul
biases
...
learning rate
−=...
'Biases' is a variable −= updates biasesSome ops compute gradients
16. Dataflow Graph
● In practice often very complex with 100s
or 1000s of nodes and edges
● “Inference” means to execute just the
forward path of the graph
17. Example: Logistic Regression
1. N examples of data
○ Input xn
○ Target tn
2. Model with parameters (weights W, biases b)
○ yn
= Softmax(W * xn
+ b)
3. CrossEntropy loss to minimize
○ loss = -Sum(tn
* ln(yn
))
4. Minimization by Gradient Descent
○ requires gradients of loss with respect to parameters W, b
● TensorFlow does automatic differentiation
● Gradient calculation done by TF once model & loss defined
28. Confidential & ProprietaryGoogle Cloud Platform 28
Easy to use API: pass in an image, we give you insight:
Google Cloud Vision API
Label
Detection OCR
Explicit
Content
Detection
Facial
Detection
Landmark
Detection
Logo
Detection