Quartz Energy - LNAM Prediction integrate with database

Challenge Objectives

 

In this challenge, we are updating an application that parses and analyzes the LAS files of a client in the oil and gas exploration industry. 

  In short, we will update app to 

• Add “Confidence” parameter of the prediction

• Store the result of prediction to MongoDb instead of current file output.

• Store the metadata of LAS file in a database as required in web UI

Project Background

Topcoder has been working with a client in the oil and gas exploration industry to develop a tool to assign a particular name (and some other data elements) to a log file.  In previous challenges, Quartz Energy - LNAM Prediction Data Science Challenge, LNAM Prediction by Curve Attribute and Vendor Name and LNAM Prediction by Curve Attribute and Vendor Name - Part 2, our community has developed an application which parses and analyzes the log files which are in LAS text format.

• This python app will be used as a service to prepare the data for a web app which yet to be developed. The data prepared by this app will be stored in the database and API will use that database to expose the endpoints to the web app.  

• With this challenge, we will update this application to integrate with the database and store any required data to develop the web app.

• After this challenge, we will create API using the database created with the challenge’s output and finally integrate with the web app. 

Technology Stack

Python 3.7, MongoDB

Code access 

Please check the forum for access to the code base. 

Individual requirements

1. Add “Confidence” parameter of the prediction

The current application produces the prediction without a confidence level. We need to show the confidence level in our algorithm's prediction for that particular LAS file. Which will be shown in the UI as 

2. Store prediction results in the database

The output of “curve-predict” command is as following format

File Name,UWI,LNAM,Service Company,Log Type,Cased Holed Flag,Generic Toolstring

The output will be saved in .csv file where filename could be passed with argument --predictions_output_file.

Now, we need to store this result in MongoDB so that this result could be shown in the above screen(Result of Predictions)

Note, there will be a new field(confidence) from 1st requirement

3. Store LAS file metadata in the database

In addition to the result of prediction, we need to store the other information related to each LAS file we processed in the database. The metadata we need to store are as shown in the following screen :

1. https://marvelapp.com/4efje50/screen/61017388
2. https://marvelapp.com/4efje50/screen/61017389
3. https://marvelapp.com/4efje50/screen/61017390
4. Parameter Info tab will have data from ~PARAMETERS INFORMATION SECTION section and data will be similar format of well info tab
5. Other Info this can be present in the LAS file or not, if it is not present on the LAS file we can ignore this.
6. For content details is the content section of the LAS file.

Let’s discuss in forum if there is anything. 

Final Submission Guidelines

Please submit the following

1. Updated source code

2. Updated readme with the deployment of database and app

3. Verification guide showing how the individual requirements are met