creating a machine learning model on the cloud

33
Building Your First Machine Learning Model With IBM Data Science Experience By Aoun Lutfi and Kunal Malhotra IBM Cloud Developer Advocates [email protected], [email protected]

Upload: alexander-al-basosi

Post on 21-Jan-2018

107 views

Category:

Software


0 download

TRANSCRIPT

Building Your First Machine Learning Model

With IBM Data Science Experience

By Aoun Lutfi and Kunal MalhotraIBM Cloud Developer [email protected], [email protected]

Agenda

1. Introduction to Data Science

2. Introduction to IBM Data Science Experience

3. Introduction to Tensorflow

4. Hands-On

IBM Confidential

3IBM Confidential

We are surrounded by, and are constantly creating

digital data. Whether it’s in emails we write, photos

we take, or where we drive; almost everything

creates data today. Data Science is the discipline of

acquiring, finding insights, and sharing discoveries in

all this data.

CRISP-DMCross Industry Standard Process for Data Mining

Concept

Methodology

Machine Learning

Neural Networks

Perceptron

Activation Function

Training – Backward Propagation

1. Initialize the weights and bias randomly.

2. Fix the input and output.

3. Forward pass the inputs. calculate the cost.

4. compute the gradients and errors.

5. Backprop and adjust the weights and bias accordingly

Convolutional Neural Networks

IBM Data Science Experience

Data Science Experience

Data Science Experience offers the opportunity to work with big data on the cloud. Use Python or R on

Spark to process big data, build models, and deploy models. Data Science Experience allows you to

easily collaborate on descriptive, prescriptive, predictive analytics, and Machine Learning on the cloud.

15

Data Science Experience

16

Data Science Experience

17

Introductionto TensorFlow

PLACE IMAGEHERE

4

TensorFlow

Originally developed by the Google Brain Team within Google'sMachine Intelligence research organisation

TensorFlow provides primitives for defining functions on tensors and automatically computing their derivatives.

An open source software library fornumerical computation using data flowgraphs

Tensor?

Simply put: Tensors can be viewed as a

multidimensional array of numbers. This means

that:

• A scalar is a tensor,

• A vector is a tensor,

• A matrix is a tensor

• ...

20

Data Flow Graph?

Computations are represented asgraphs:

• Nodes are the operations(ops)

• Edges are theTensors (multidimensional arrays)

Typicalprogram consists of 2 phases:

• construction phase: assembling a graph (model)

• execution phase: pushing data through thegraph

21

Neural Networks? DeepLearning?

22

● Neural Networks are represented by the lower figure, not the

top one....

● Link:

Tinker with a Neural Network inYour Browser

Presentation title (Go to View > Master to edit) 8Source: https://www.udacity.com/course/deep-learning--ud730

Presentation title (Go to View > Master to edit) 9Source: https://www.udacity.com/course/deep-learning--ud730

Presentation title (Go to View > Master to edit) 15Source: https://www.udacity.com/course/deep-learning--ud730

Presentation title (Go to View > Master to edit) 16Source: https://www.udacity.com/course/deep-learning--ud730

18

Why would you use NN / Deep Learning?

• Neural Networks (NNs) are universal function

approximators that work very well with huge

datasets

• NNs / deep networks do unsupervised feature

learning

• Track record, being SotA in:

• image classification,

• language processing,

• speech recognition,

• ...

19

Why TensorFlow?

There are a lot of alternatives:

● Torch

● Caffe

● Theano (Keras, Lasagne)

● CuDNN

● Mxnet

● DSSTNE

● DL4J

● DIANNE

● Etc.

20

TensorFlow has the largestcommunity

Sources: http://deliprao.com/archives/168

http://www.slideshare.net/JenAman/large-scale-deep-learning-wit

h-tensorflow

Runs on CPUs, GPUs, TPUs over one or more

machines, but also on phones(android+iOS) and

raspberrypi’s...

TensorFlow is very portable/scalable

30

TensorFlow is more than an R&D project

• Specific functionalities for deployment (TF Serving /

CloudML)

• Easier/more documentation (for more general public)

• Included visualization tool(Tensorboard)

• Simplified interfaces likeSKFlow

31

32

Hands On Lab

Building your first Machine Learning model on IBM Data Science Experience.

Sign in to IBM Cloud on: ibm.biz/Intro2MLonDSX

Access Data Science Experience on: datascience.ibm.com

GitHub Link: github.com/aounlutfi/building-first-ML-model

MNIST CNN using

Tensorflow

Build an CNN to classify handwritten digits using the MNIST dataset.