Overview

All of the REST APIs call presented below use bearer tokens for authorization. The {prefix} of each API is configurable in the hosted servers. This protocol is inspired by Delta Sharing.

Protocol - Training

* Blue text below, either in the diagram or in the description, means it's out of scope of OBAIC and it's up to the BI tool, AI Platform or Data source vendor to implement. OBAIC is the connecting tissue to coordinate the communications among them to extend the capability of these 3 major components


  1. User analyzes data using BI Tools and found out that predictive analytics on those data set would be valuable. This step is the traditional step when a user interacts with BI.
  2. (a) Obtain a token a token with permission associated to the user making the request. This token is going to pass to AI allowing the access to the training data with a SQL statement running against the datastore. (b) BI tool, on behalf of the user, requests AI platform through OBAIC, to train/prepare a model that accepts features of a certain type (numeric, categorical, text, etc.)

    Model configuration is based on configs from the open-source Ludwig project. At a minimum, we should be able to define inputs and outputs in a fairly standard way. Other model configuration parameters are subsumed by the options field.

    The data stanza provides a bearer token allowing the ML provider to access the required data table(s) for training. The provided SQL query indicates how the training data should be extracted from the source.

    Don't be confused with the Bearer token which is used to authenticate with OBAIC, and the dbToken which is created in 2(a) and AI platform will use that to access the data source for training

    HTTP RequestValue
    Method

    POST

    Header

    Authorization: Bearer {token}

    URL

    {prefix}/models/

    Query Parameters

    {

      "dbToken": "D41C4A382C27A4B5DF824E2D4F148";
      "inputs":[
        {
          "name":"customerAge",
          "type":"numeric"
        },
        {
          "name":"activeInLastMonth",
          "type":"binary"
        }
      ],
      "outputs":[
        {
          "name":"canceledMembership",
          "type":"binary"
        }
      ],
      "modelOptions": {

          “providerSpecificOption”: “value”

       },
      "data":{
          "sourceType":"snowflake",
          "endpoint":"some/endpoint",
          "bearerToken":"...",
          "query":"SELECT foo FROM bar WHERE baz"
      }
    }



    If we go beyond just REST API, SQL-like is an alternative as the syntax is also well-known

    Use BigQuery ML model creation as an example and generalizing

    CREATE MODEL (
      customerAge WITH ENCODING (
        type=numeric
      ),
      activeInLastMonth WITH ENCODING (
        type=binary
      ),
      canceledMembership WITH DECODING (
        type=binary
      )
    )
    FROM myData (
      sourceType=snowflake,
      endpoint="some/endpoint",
      bearerToken=<...>,

    AS (SELECT foo FROM BAR)
    WITH OPTIONS ();




    HTTP ResponseValue
    HeaderContent-Type: application/json; charset=utf-8
    Body

    {
      "modelID": "d677b054-2cd4-4711-959b-971af0081a73"
    }

    • modelID is generated and returned to the caller if training is started successfully. This will be used to check the status of the training, or for future Inference (see Inference section below)



  3. AI Platform provider the implementation to fulfill the request by connecting to the datasource with the provided token and the set of training data specified in SQL. This step is up to how the AI platform interacts with the data source to performance the training. 
  4. BI tool polls for the status or retrieve the training result. If the training is still in progress, the status will be returned. When training is completed, results and performance of the model will be returned.


    HTTP RequestValue
    Method

    GET

    Header

    Authorization: Bearer {token}

    URL

    {prefix}/modelStatus?modelID=

    Query Parameters

    modelID (type: String): The modelID returned from previous OBAIC call either from training or list of Models.




    HTTP ResponseValue
    HeaderContent-Type: application/json; charset=utf-8
    Body

    {
      "modelID": "d677b054-2cd4-4711-959b-971af0081a73",

      "status": "training",

      "progress": "80",
    }

    • modelID is same ID provided in the request
    • status can be training | inferencing | ready
    • progress is the estimated progress of the current status



  5. BI tool presents the result to the user in their own way, which is the "secret sauce" and unique to each other.

Protocol - Inference

1. When BI wants to extend its capability to AI, it reaches out to AI platform and request for a list of available models of which the credential of the provided token is authorized to see


HTTP RequestValue
Method

GET

Header

Authorization: Bearer {token}

URL

{prefix}/models/{model}

Query Parameters

maxResults (type: Int32, optional): The maximum number of results per page that should be returned. If the number of available results is larger than maxResult, the response will provide a nextPageToken that can be used to get the next page of results in subsequent list requests. The server may return fewer than maxResults items even if there are more available. The client should check nextPageToken in the response to determine if there are more available. Must be non-negative. 0 will return no results but nextPageToken may be populated.

pageToken (type: String, optional): Specifies a page token to use. Set pageToken to the nextPageToken returned by a previous list request to get the next page of results. nextPageToken will not be returned in a response if there are no more results available.




HTTP ResponseValue
HeaderContent-Type: application/json; charset=utf-8
Body

{
  "items": [
    {
      "name": "string",
      "id": "string"
    }
  ],
  "nextPageToken": "string"
}

  • items will be an empty array when no results are found. 
  • id field is the key to retrieved the model in the subsequent calls. Its value must be unique across the AI server and immutable through the model's lifecycle.
  • nextPageToken will be missing when there are no additional results


Example:

{
   "models": [
      {
         "name": "Model 1",
         "id": "6d4b571a-80ca-41ef-bc67-b158f4352ad8"   
      },
      {
         "name": "Model 2",
         "id": "70d9ab9d-9a64-49a8-be4d-d3a678b4ab16"
      },
      {
         "name": "Model 3",
         "id": "99914a97-5d2e-4b9f-b81a-1d43c9409162"
      },
      {
         "name": "Model 4",
         "id": "8295bfda-7901-43e8-9d31-81fd1c3210ee"
      },
      {
         "name": "Model 5",
         "id": "0693c224-3a3f-4fe7-bbbe-c70f93d15f12"
      }
   ],
   "nextPageToken": "3xXc4ZAsqZQwgejt"
}

2. After the list of models is return, BI can selectively retrieve the detail of the model. This step can also be called right after the newly trained model is completed as described in the previous section since modelID is returned as a result of the training request.


HTTP RequestValue
MethodGET
HeaderAuthorization: Bearer {token}
URL{prefix}/model/{modelID}
URL Parameters{modelID}: The case-insensitive ID of the model returned in in List Models for Step (1)




HTTP ResponseValue
Header

Content-Type: application/json; charset=utf-8

Body

{
  "id": "string",
  "name": "string",
  "revision": "int",
  "format": {
    "name": "string", 
    "version": "string"
  },
  "algorithm": "string", // Artificial neural network | Decision trees | Support-vector machines | Regression analysis | Bayesian networks | Genetic algorithms | Proprietary
  "tags": ["string"],
  "dependency": "string",
  "creator": "string",
  "description": "string",
  "input": {
    "fields": [
      {
        "name": "string",
        "opType": "string",
        "dataType": "string",
        "taxonomy": "string",
        "example": "string",
        "allowMissing": "boolean",
        "description": "string"
      }, ...
    ],
    "$ref": "string"
  },
  "output": {
    "fields": [
      {
        "name": "string",
        "opType": "string",
        "dataType": "string",
        "taxonomy": "string",
        "example": "string",
        "allowMissing": "boolean",
        "description": "string"
      }, ...
    ],
    "$ref": "string"
  },
  "performance": {
    "metric": "string",
    "value": "float"
  },
  "rating": "int",
  "url": "string"
}

  • format.name: PMML, ONNX, or other formats to be confirmed
  • algorithm: Artificial neural network | Decision trees | Support-vector machines | Regression analysis | Bayesian networks | Genetic algorithms | Proprietary
  • tags: describe what this model is used for e.g. Agriculture | Banking | Computer vision | Credit-card fraud detection | Handwriting recognition | Insurance | Machine translation | Marketing | Natural language processing | Online advertising | Recommender systems | Sentiment analysis | Telecommunication | Time-series forecasting | etc.
  • opType: categorical | ordinal | continuous
  • dataType: string | integer | float | double | boolean | date | time | dateTime
  • $ref: reference to external schema for the format used
  • metric: based on model used, metric can be accuracy, precision, recall, ROC, AUC, Gini coefficient, Log loss, F1 score, MAE, MSE, etc.
  • url: link to the real model for download


Example:

{
    "id": "6d4b571a-80ca-41ef-bc67-b158f4352ad8",
    "name": "Model 1",
    "revision": 3,
    "format": { 
      "name": "PMML",
      "version": "4.3"
    },
    "algorithm": "Neural Network", 
    "tags": [
      "Anomaly detection",         
      "Banking"                    
    ],                              
    "dependency", "",
    "creator": "John Doe",
    "description": "This is a predictive model, refer to {input} and {output} for detailed format of each field, such as value range of a field, as well as possible predictions the model will gave. You may also refer to the example data here.",
    "input": {
      "fields": [
        {
          "name": "Account ID",
          "opType": "categorical",
          "dataType": "string",
          "taxonomy": "ID",
          "example": "account abc-001",
          "allowMissing": false,
          "description": "unique value"
        },
        {
          "name": "Account Balance",
          "opType": "continuous",
          "dataType": "double",
          "taxonomy": "currency",
          "example": "1,378,560.00",
          "allowMissing": true,
          "description": "Minimum: 0, Maximum: 999,999,999.00"
        }, 
      ],
      "ref": "http://dmg.org/pmml/v4-3/pmml-4-3.xsd"                                                       
    }
    "output": {
      "fields": [
        {
          "name": "Churn",
          "opType": "continuous",
          "dataType": "string",
          "taxonomy": "ID",
          "example": "0.67",
          "allowMissing": false,
          "description": "the possibility of the account stop doing business with a company over 6 months"
        }
      ],
      "ref": "http://dmg.org/pmml/v4-3/pmml-4-3.xsd"                                                       
    }
    "performance": {            
      "metric": "accuracy",     
      "value": 0.85
    },
    "rating": 5,
    "url": "uri://link_to_the_model"  
}

3. BI tool will use the information retrieved from the AI platform to display to the user, including what type of models are available and the performance. It can optionally match the data and suggest what may be the good match based on what the user has.

4. User interacts with the result BI presented and decides what can be a good model to make a prediction on certain set of data. Please note that the model can also be returned as the result of the training step described in the previous section. In the case, the user may bypass these 2 steps and go directly to see the result.

5.

Error - Apply to all API calls above


HTTP ResponseValue
HeaderContent-Type: application/json
Body{
"errorCode": "string",
"message": "string"
}




HTTP ResponseValue
HeaderContent-Type: application/json
Body{
"errorCode": "string",
"message": "string"
}




HTTP ResponseValue
HeaderContent-Type: application/json
Body{
"errorCode": "string",
"message": "string"
}




HTTP ResponseValue
HeaderContent-Type: application/json
Body{
"errorCode": "string",
"message": "string"
}


Next Step

Future Enhancement

FAQ

  1. Why should AI share model to BI?
  2. Who owns the model and data?
  3. How do you deal with Security?

References

Authors

Name

Affiliation

Cupid ChanPistevo Decision
Xiangxiang MengRedfin
Deepak KaruppiahMicroStrategy
Nancy RauschSAS
Dalton RuerQlik
Sachin SinhaMicrosoft
Yi ShaoIBM
Jeffrey TangPredibaes
Lingyan YinSalesforce



Get Model Metrics

Get core evaluation metrics for a trained model.

function GetModelMetrics(UUID) -> Metrics


Example response:

{
  "accuracy":0.781,
  "lossType":"cross-entropy",
  "loss":0.0238
}


Predict Using Trained Model

function PredictWithModel(UUID, dataConfig) -> Predictions


Example params

{
  "uuid": "abcdef12345",
  "data":{
      "sourceType":"snowflake",
      "endpoint":"some/endpoint",
      "bearerToken":"...",
      "query":"SELECT foo FROM bar WHERE baz"
  }
}

    

A very similar data stanza to the train request, designating the feature data on which to predict.

Example response (as JSON here for convenience, not necessarily for large responses):

{
  "data":[
      {
        "customerAge":2,
        "activeInLastMonth":"false",
        "predicted__canceledSubscription":"true"
      },
      {
        "customerAge":9,
        "activeInLastMonth":"true",
        "predicted__canceledSubscription":"false"
      }
  ]
}


Note that directly returning a large response set is not a good idea. In practice, the results could be streamed through something like a persistent socket connection.