Bluemix_RR_Nodejs

Bluemix API of retrieve and rank in Node.js

What is this repository for?

• Quick summary: Use retrieve and rank API on Bluemix by Node.js
• Note: This is for Chinese version of RR, Solr_Japanese_Chi_20160509 and train_with_encoded_ja are specific for this purpose.
• Version 1.1.0

Preparation

1. Apply a Bluemix account.
2. Be familiar with Nodejs fundamentation.

Stage One - Set up service

1. Log in at Bluemix website.
2. Go to catalog and find the R&R service.
3. Type a unique name for the service instance in the Service name field. Leave the default values for the other options.
4. In your Service Credentials, bear in mind your username and password.

Stage Two - Create, train and rank a classifier

1. Go to Nodejs SDK from Github page which is developed by IBM.
2. Open get started [page[(https://www.ibm.com/watson/developercloud/doc/retrieve-rank/index.shtml) to download some resources.
3. Install node-nodule:$ npm install watson-developer-cloud --save

• Credential:

  var retrieve_and_rank = watson.retrieve_and_rank({
  username: '<username>',
  password: '<password>',
  version: 'v1'
  });

• Create a cluster first.

  function createCluster() {
  retrieve_and_rank.createCluster({
  cluster_size: '1',
  cluster_name: 'API_version'
  },
  function (err, response) {
    if (err)
      console.log('error:', err);
    else
      console.log(JSON.stringify(response, null, 2));
  });
  }

• Upload configuration file

  function uploadConfig() {
  
  var params = {
  cluster_id: cluster_id,
  config_name: config_name,
  config_zip_path: 'Solr_Japanese_Chi_20160509.zip'
  };
  
  retrieve_and_rank.uploadConfig(params,
  function (err, response) {
    if (err)
      console.log('error:', err);
    else
      console.log(JSON.stringify(response, null, 2));
  });
  }

• Create a collection

  function createCollection(name) {
  var params = {
  cluster_id: cluster_id,
  config_name: config_name,
  collection_name: name,
  wt: 'json'
  };
  
  retrieve_and_rank.createCollection(params,
  function (err, response) {
    if (err)
      console.log('error:', err);
    else
      console.log(JSON.stringify(response, null, 2));
  });
  }

• Upload documents.

  function uploadDoc(path) {
  
  var file = path;
  var doc = jsonfile.readFileSync(file);
  
  solrClient = retrieve_and_rank.createSolrClient({
  cluster_id: cluster_id,
  collection_name: collection_name,
  });
  
  console.log('Indexing a document...');
  solrClient.add(doc, function (err, response) {
  if (err) {
    console.log('Error indexing document: ', err);
  }
    else {
      console.log('Indexed a document.');
      solrClient.commit(function(err) {
        if(err) {
          console.log('Error committing change: ' + err);
        }
          else {
            console.log('Successfully committed changes.');
          }
      });
    }
  });
  }

• Create and train the ranker. Please DON’T use node,js or Curl this step. Just use Python file to process.

  $ python ./train_with_encoded_ja.py -u "<username>":"<password>"

• Search and Rank

  function searching(input) {
  
  solrClient = retrieve_and_rank.createSolrClient({
  cluster_id: cluster_id,
  collection_name: collection_name,
  wt: 'json'
  });
  
  var ranker_id = '';
  var question  = 'q=' + input;
  var query     = qs.stringify({q: question, ranker_id: ranker_id, fl: 'id,body'});
  
  solrClient.get('fcselect', query, function(err, searchResponse) {
  if(err) {
    console.log('Error searching for documents: ' + err);
  }
    else {
      var result = JSON.stringify(searchResponse.response.docs, null, 2);
      result = JSON.parse(result);
      if (result == '') {
        console.log('Not found');
      } else {
        console.log(result);
      }
  
    }
  });
  }

How do I get set up?

0. Read IBM Watson Developer Cloud to know concept and flow. Sign up/in on Bluemix and crate a Retreive and Rank service with usr/pwd

1. Install nodejs on your computer: https://nodejs.org/en/

2. Oepn your Terminal/Command Line Tool.

3. go to Project Path: cd /path

4. Install modeules: npm install (if you encouter permission problems: sudo npm install)

5. Main functions are all implement in Solrcluster.js file.

   • Create Cluster.
   • Create collection.
   • Upload configuration.
   • Upload Documents.
   • Upload Training data and create Ranker.
   • Search and rerank.
   • Get information about configuration/cluster/rank.

6. Enjoy it (Ask something).

7. Remember Close the sever by Ctl+c

How to run program

• Use CMD/Terminal
• Web browser

Contribution guidelines

• IBM Watson Developer Cloud
• Bo Cheng Huang