Skip to content

BluemixCloud/2016-09-26

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

11 Commits
images
images
 
 
README.md
README.md
 
 
openwhisk.md
openwhisk.md
 
 

Repository files navigation

IBM Cloud Advisor 101 Course

Galvanize Watson IBM Bluemix

Cohort 2016-09-26

Food Schedule

  • Monday: 8am - 5pm (Breakfast and Lunch served)
  • Tuesday - Friday: 9am - 5pm
  • Friday: 5pm (Hosted Happyhour on Rooftop)

Coding Schedule

  • Monday: Fundamentals, Backend
  • Tuesday: DevOps, Cloud Foundry, OpenWhisk, NodeRED
  • Wednesday: Fullstack Apps with Microservices
  • Thursday: Fullstack Apps with Microservices
  • Friday: Hackathon

Syllabus

Arrive & Setup

  • Meet & Greet + Breakfast
  • Get on Network (g|events -> machinelearning)

Introductions

Bluemix Garage

Assessment

  • Technical
  • Cloud
  • Agile

Cloud 9 Setup

  • Register for edX (free)
  • https://www.edx.org/
  • They will send you a confirmation email. Click link to activate.
  • Go to the following URL to create your Cloud 9 account. Login using your edX account.
  • https://cs50.io/
  • After you have a running virtual machine, go to your dashboard, which is the URL + your username
  • https://cs50.io/USERNAME
  • Create a Custom Ubuntu Workspace
  • Open Terminal
wget https://raw.githubusercontent.com/chyld/devops/master/c9-setup.sh
chmod +x c9-setup.sh
./c9-setup.sh
vi ~/.gitconfig
exit

Git & JazzHub

JazzHub

JazzHub

Git Commands

# To clone an existing repository
git clone https://github.com/BluemixCloud/2016-03-28

# To initialize an empty git repository
git init

# To check the status of your repository
git status

# To see all the local and remote branches
git branch -av

# To check your remote repository
git remote -v

# To see all of your commits
git log

# To push your code up to the remote repository
git add .
git commit -am "enter a commit message here"
git pull
git push

Modern Programming Fundamentals

JS Runtime

Project A

  • This project will be done as a pair
  • Pick a project manager
  • The PM will go here, https://hub.jazz.net/project/chyld/fundamentals
  • PM, click the "Fork Project" button
    • Name the project: fundamentals
    • Check all the boxes
    • Click Create
  • PM, click the "Members" link on the left, add your partner to the repository
  • Both people can now push and pull the repository
  • Both, on the project page, click the "Git URL" link
  • Both, copy that link
  • Both, open C9 terminal
  • Both, make sure you're in the ~/workspace directory
git clone <link you copied above>
cd fundamentals
npm install
npm test test/square.js # this should pass
npm test test/add.js # this fails, fix
npm test # to run all tests
  • Fix the remaining tests
  • Take turns, with both people fixing the failing tests
  • Push your completed code up the repository

Backend

DevOps Services

  • https://hub.jazz.net/
  • JazzHub
  • Code Editor
  • Track & Plan
  • Build Pipeline
    • Use manifest.yml from Backend
    • Build a Stage
    • Add Jobs to a Stage
    • Jobs can be re-ordered

Cloud Foundry

Project B

Chrome Plugins

Front End Technologies

Project C

OpenWhisk

NodeRED

Project 2

From Angular send a message to Express then Twilio and finally to your Phone

Steps

  1. Stamplay: Make a webhook, call it todo. Test the webhook with http client.
  2. Twilio: Edit your Phone Number. Change messaging to post to the Stamplay webhook from Step 1.
  3. Send a text message to Twilio. The data should appear in Stamplay debugger.
  4. Trello: Create a Board
  5. Trello: Create a List
  6. Stamplay: Create a Task. The Task will wire up the webhook and the Trello board
  7. Send a text message. The message should appear in your Trello board list.

Project 3

Display the Top Story from ANY Reddit Topic and Display in Slack

Steps

  1. If you are not an Administrator on Slack, Create Your Team Account
  2. Sign into your Slack account and create a channel
  3. NodeRED: Create an post /reddit. Test from http client.
  4. Create an Outgoing Webhook on Slack. Use the reddit as the trigger word
  5. Send a message from your slack channel to NodeRED. You should have debugging info displayed.
  6. Create an OpenWhisk action. Deploy and Test it.
var Request = require('request');

function main(params){
  var channel = params.channel;
  var url = 'https://www.reddit.com/' + channel + '.json';
  Request({url: url, json: true}, function(e, r, b){
    console.log('error:', e);
    console.log('status:', r.statusCode);
    var first = b.data.children[0].data;
    whisk.done({payload: {title: first.title, url: first.url}});
  });

  return whisk.async();
}
  1. NodeRED: Prepare the data before the call to OpenWhisk.
var channel = msg.payload.text;

channel = channel.replace(/reddit/g, '');
channel = channel.trim();

msg.payload = {channel: channel};

return msg;
  1. NodeRED: After the call to OpenWhisk, prepare the data to be sent back to your slack channel.
var data = msg.payload.payload;

msg.payload = {
    text: data.title + ' ' + data.url
};

return msg;

Project 4

A user fills out a customer response form. The form gets analyzed by Watson Tone Analyzer with the results streamed in realtime to your slack channel.

Steps

  1. Create a form on WuFoo
  2. The form should have two fields, a name field and feedback field
  3. Get the subdomain name and API key for your form
  4. Go to Stamplay, Under Tasks -> Components
  5. Connect to Slack and WuFoo
  6. Make a Code Block called wufoo
module.exports = function(context, cb) {
  var Request = require('request');
  var data = context.data;
  var url = 'http://_____your_node_red.mybluemix.net/wufoo';

  if(data.data){
	data = data.data;
    data = JSON.parse(data.replace(/\n|\r/g, ''));
  }

  console.log('data:', data);

  Request({url: url, json: true, method: 'post', body: data}, function(e, r, b){
	cb(null, { payload : b});
  });
} ;
  1. Create a Task
  2. WuFoo -> Code Block
  3. Send data to the Code Block
{
"name" : "{{entry.body.Field8}}",
"feedback" : "{{entry.body.Field4}}"
}
  1. Create a simple post /wufoo endpoint in NodeRED
  2. Fill out the form and look at the deugging info
  3. You may need to change the code in step 9
  4. Go back to NodeRED and add a prep function before the Tone Analyzer
msg.save = msg.payload;
msg.payload = msg.save.feedback;
return msg;
  1. Add a Tone Analyzer node
  2. Tones -> Emotion, Sentences -> True
  3. Add some post processing after the Tone Analyzer, before the data gets sent back to Stamplay
var tones = msg.response.document_tone.tone_categories[0].tones;

tones = tones.map(function(t){
    return t.tone_name + ': ' + (t.score * 100).toFixed(2) + '%';
});

var text = msg.save.name + ' -- ' + tones.join(', ');

msg.payload = {text: text};

return msg;
  1. Add some debugging nodes. Make sure all looks good.
  2. Go back to Stamplay. Add another Task.
  3. Code Block -> Slack
  4. Have your wufoo Code Block, upon execution, insert a message into a Slack channel that you choose.
  5. Test everything out

Project 5

  • Twitter Sentiment Analysis
  • Prerequisites
  • Fork and Clone the Fullstack Template
  • Use NodeRED to create a flow that collects data from Twitter, performs Sentiment Analyis and inserts all that data into a DB2 SQL Database.
  • Create another flow to query the database and send back the result as JSON.
  • Create a button in Angular that will fetch all the data and display the data in a table and graph.

Code that aggregates the tweet + sentiment analysis data; ready for insertion into DB2.

var place;

if(msg.location){
    place = msg.location.place;
}

msg.payload = {
    SCORE: msg.sentiment.score,
    TWEET: msg.tweet.text,
    USERNAME: msg.tweet.user.screen_name,
    LOCATION: place || 'unknown',
    CREATED_AT: 'TIMESTAMP'
};

return msg;

Script that creates the table inside DB2. Where the tweet data will be stored.

create table tweets
(
  id integer not null generated always as identity (start with 1 increment by 1),
  primary key (id),
  score integer,
  tweet varchar(256),
  username varchar(256),
  location varchar(256),
  created_at timestamp
);

SQL to collect all the tweet data from DB2.

select * from tweets;

Angular code to draw a chart using Highcharts.

function drawChart(tweets){
  $('#graph').highcharts({
    title: {
      text: 'Tweet Sentiment Analysis'
    },
    xAxis: {
       categories: tweets.map(function(t, i){return i})
    },
    series: [
     {
       data: tweets.map(function(t){return t.SCORE})
     }
    ]
  });
}

Project 6

Project 7

Project 8

  • Camera Image Capture & Watson Analyis
  • Prerequisites
  • Fork and Clone the Fullstack Template
  • User snaps a photo of herself using the laptop's webcam.
  • The photo gets uploaded to Object Storage.
  • Then the photo gets pulled into NodeRED, gets analyzed by Watson Visual Recognition
  • That data gets stored in a Cloudant database
  • A browser will query the Cloudant database and display all the results

Project 9

Audio

Whiteboard

Buttons inside HTML

<button ng-click="startRec()">Start</button>
<button ng-click="stopRec()">Stop</button>

Place at the bottom of your angular code in index.js

var audio_context;
var recorder;

initAudio();

function startUserMedia(stream) {
  var input = audio_context.createMediaStreamSource(stream);
  recorder = new Recorder(input);
}

function initAudio(){
  try {
    window.AudioContext = window.AudioContext || window.webkitAudioContext;
    navigator.getUserMedia = navigator.getUserMedia || navigator.webkitGetUserMedia || navigator.mozGetUserMedia;
    window.URL = window.URL || window.webkitURL;

    audio_context = new AudioContext;
  } catch (e) {
    alert('No web audio support in this browser!');
  }

  navigator.getUserMedia({audio: true}, startUserMedia, function(e) {
    console.log('No live audio input: ' + e);
  });
}

function uploadAudio(){
  recorder && recorder.exportWAV(function(blob) {
    var formData = new FormData();
    formData.append("voice", blob);
    var request = new XMLHttpRequest();
    request.open("post", "/audio");
    request.send(formData);
  });
}

Functions to start and stop recording - also upload to server

 $scope.startRec = function(){
    console.log('Recording: Start');
    recorder && recorder.record();
  };

  $scope.stopRec = function(){
    console.log('Recording: Stop');
    recorder && recorder.stop();
    uploadAudio();
    recorder.clear();
  };

REST API Endpoint in Express.js that will upload audio to NodeRED

var os = new ObjectStorage('user_id', 'password', 'project_id', 'container_name', 'access_point_url');
os.setContainerPublicReadable();

app.post('/audio', uploadr.single('voice'), function(req, res){
  var name = 'audio-original-' + uuid.v1() + '.wav';
  os.uploadFileToContainer(name, req.file.mimetype, req.file.buffer, req.file.size)
  .then(function(file){
    var o = {url: 'http://___your_node_red___.mybluemix.net/translate-audio', json: true, method: 'post', body: {url: file}};
    request(o, function(e, r, b){
      res.send({payload: 'success'});
    });
  });
});

Hackathon

Retrospective

Feedback

Tour

  • Galvanize
  • Full Stack Classrooms
  • Data Science Classrooms
  • Bluemix Garage
  • Cloudant
  • Startups

Happy Hour on Roof

Collect Badges

REFERENCES

Bluemix Services

Books

Reference Links

About

No description, website, or topics provided.

Resources

Readme
Activity
Custom properties

Stars

0 stars

Watchers

2 watching

Forks

1 fork
Report repository

Releases

No releases published

Packages

No packages published