Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
Getting Started with TensorFlow
Search
Rebecca Murphy
March 21, 2016
Programming
0
1.4k
Getting Started with TensorFlow
Rebecca Murphy
March 21, 2016
Tweet
Share
More Decks by Rebecca Murphy
See All by Rebecca Murphy
Refreerank
rebecca_roisin
0
120
pyFRET
rebecca_roisin
0
220
Other Decks in Programming
See All in Programming
プロパティベーステストによるUIテスト: LLMによるプロパティ定義生成でエッジケースを捉える
tetta_pdnt
0
4.3k
Processing Gem ベースの、2D レトロゲームエンジンの開発
tokujiros
2
130
Deep Dive into Kotlin Flow
jmatsu
1
370
Updates on MLS on Ruby (and maybe more)
sylph01
1
180
Platformに“ちょうどいい”責務ってどこ? 関心の熱さにあわせて考える、責務分担のプラクティス
estie
1
140
2025 年のコーディングエージェントの現在地とエンジニアの仕事の変化について
azukiazusa1
24
12k
もうちょっといいRubyプロファイラを作りたい (2025)
osyoyu
1
460
Performance for Conversion! 分散トレーシングでボトルネックを 特定せよ
inetand
0
3.4k
詳解!defer panic recover のしくみ / Understanding defer, panic, and recover
convto
0
250
OSS開発者という働き方
andpad
5
1.7k
GitHubとGitLabとAWS CodePipelineでCI/CDを組み比べてみた
satoshi256kbyte
4
250
CJK and Unicode From a PHP Committer
youkidearitai
PRO
0
110
Featured
See All Featured
BBQ
matthewcrist
89
9.8k
The Success of Rails: Ensuring Growth for the Next 100 Years
eileencodes
46
7.6k
Dealing with People You Can't Stand - Big Design 2015
cassininazir
367
27k
Embracing the Ebb and Flow
colly
87
4.8k
The Cult of Friendly URLs
andyhume
79
6.6k
Reflections from 52 weeks, 52 projects
jeffersonlam
352
21k
Large-scale JavaScript Application Architecture
addyosmani
513
110k
Visualization
eitanlees
148
16k
Java REST API Framework Comparison - PWX 2021
mraible
33
8.8k
YesSQL, Process and Tooling at Scale
rocio
173
14k
Raft: Consensus for Rubyists
vanstee
140
7.1k
Gamification - CAS2011
davidbonilla
81
5.4k
Transcript
TensorFlow Tutorial Rebecca Murphy
[email protected]
@rebecca_roisin TensorFlow Meetup Monday 21st
March 2016
Talk Overview • TensorFlow overview • Programming Model • Mechanics
of TensorFlow • Installation • Model Definition • Fitting • Checkpointing • TensorBoard visualisations • Why TensorFlow?
TensorFlow: Overview
What Is TensorFlow? • Google’s 2nd generation deep learning library
• Simple API (Python, C++) for: • Describing Machine Learning models • Implementing Machine Learning algorithms
What Can We Do With TensorFlow? • Regression models •
Neural networks • Deep learning: • Distributed representations • Convolutional Networks • Recurrent Neural Networks • LSTM Neural Networks
TensorFlow: Programming Model
TensorFlow: What Is a Tensor? • Tensor: n-dimensional array •
Scalar: 0D Tensor • Vector: 1D Tensor • Matrix: 2D Tensor • Typed: • Int, double, complex, string
Tensor Flows • Tensor Flow computations: stateful dataflow graphs •
Deep learning model = Directed graph • Node: (mathematical) operation • Edge: • Control dependencies • Data flow • Describe graph -> initialize -> execute (parts of ) graph
TensorFlow: Mechanics
Installing TensorFlow • Python API • Python 2.7 • Python
3.3+ • Setup instructions • pip install: • pip install --upgrade https://storage.googleapis. com/tensorflow/mac/tensorflow-0.7.1-cp27-none-any.whl • Docker: • docker run -it b.gcr.io/tensorflow/tensorflow
Mechanics of Learning • Define model • Load data •
Feed data • Make predictions • Evaluate • Visualise
Example Code • Try-tf github repositories • Associated blogpost •
Jason Baldridge @jasonbaldridge
Let’s get Started
Defining the Model
Model Definition: Key Features (1) • Tensor shapes are pre-defined:
• Tensors support mathematical manipulation • Operations are nodes in the model graph
Model Definition: Key Features (2) • Built-in functions for common
Deep Learning operations: • See Neural Network API for more • Gradient descent optimisation: • Variables store current state of model
Training the Model: Loading Data (1) • Load data into
variables • Need to write custom functions to parse data
Training the Model: Loading Data (2)
Training the Model: Sessions • Model graph describes computations •
Computations evaluated within a session: • Places graph onto CPU / GPU • Supplies methods to evaluate graph operations
The Feed Dict: Training the Model • Predefined placeholder tensors
• Feed-dict supplies batch of data
Training the Model: Evaluation • Pre-defined evaluation nodes compare predicted
and true labels: • Evaluate accuracy function within a session:
Checkpoints: Saving Models • Saver class allows model state to
be stored and reloaded • Use checkpoints to periodically save the state of the model
• Saver class allows model state to be stored and
reloaded • Restore a previously trained model Checkpoints: Loading Saved Models
Flags: Controlling Training • tf.app.flags: set command-line arguments • Wraps
python gflags • tf.app.run() parses flags before calling main()
TensorBoard: Visualising Learning
TensorBoard: Basics • TensorFlow visualisation tool • View • Graph
models • Training behaviour • Simple modifications to model code • Browser-based tool
TensorBoard: Annotations
TensorBoard: Scopes
TensorBoard: Saving Output • Set up summary and writer objects
• Periodically run evaluation and store output: • tensorboard --logdir=try_tf_logs/
TensorBoard: Model Visualisation (1)
TensorBoard: Model Visualisation (2)
TensorBoard: Training Visualisation (1)
TensorBoard: Training Visualisation (2)
TensorFlow: Where Next?
Why Use TensorFlow: Great Examples • TensorFlow Tutorials • Handwriting
generation from @hardmaru • Next letter prediction from @karpathy
Why Use TensorFlow: Active Community
TensorFlow: Future Developments • Improved memory usage in gradient calculations
• JIT Compilation • Improved node execution scheduling • Support for parallelisation across many machines • Support for more languages (Java, Lua, Go, R …) • Source: TensorFlow Whitepaper
Thank You!
Questions?