Predigt.dev

Predigt.dev

Predictive Model: My First Tech Task

Without knowledge of something, the application will be an uphill struggle

Predictive Model: My First Tech Task

Subscribe to my newsletter and never miss my upcoming articles

Listen to this article

During my early days of transitioning to tech, I took my first step in learning Data Science and Analytics on IBM Digital-Nation Africa. One of the modules has a task to create a Predictive Model-Analyzing and Predicting Heart Failure on IBM Cloud. I have worked on many non-tech-related projects in my career, the application task caught my eye because each year hundreds of thousands of people die from heart disease and stroke, of which most of the major risk factors can be managed or prevented.

790510ED-C1C4-4F39-8ECD-9E0B8724E12E.jpeg

It was challenging as it is my first time working on somewhat a tech project, so I gave it a shot because it is also a chance to confirm if I’m on the right path. For a novice with little or no coding experience, nor expertise in many technology domains, it wasn’t an easy task although steps and code to create and deploy the model were provided.

In the process I came across some technologies like Artificial Intelligence and Node.js and used components like;

Notebooks:

Jupyter notebook a web-based environment for interactive computing. You can run small pieces of code that process your data, and immediately view the results of your computation.

PixieDust:

An open-source productivity tool for Jupyter Notebooks, it helps to visualize data in a way that’s easy to share and consume and provides a Python helper library for IPython Notebook.

Spark:

Apache Spark is a highly versatile, open-source cluster computing framework with fast, in-memory analytics performance.

Watson Studio:

To analyze data using RStudio, Jupyter, and Python in a configured, collaborative environment that includes IBM value-adds, such as managed Spark. Watson Machine Learning Service for hosting your trained model on IBM’s Cloud.

These were part of my key challenge, as the tools were somewhat complex for a total beginner.

Three times I worked on the task, and 3 times the test failed. I deleted the project, stayed up late at night, asked for help from the community, redo the entire task, and eventually brought the project to completion.

6F0EDFE1-C560-4128-9F3A-81BD1085671F.jpeg

246E1CB2-F74B-4574-BA4F-15C2C2C57CFA.jpeg

During the phase, I acquired, clean, and explore data, build a predictive machine learning model, make predictions, host the model on IBM Cloud for consumption and use and Integrate the model with your web application.

Asides from getting to use a new tool, I have learned that it is more than just meeting a goal, it is about having the knowledge or right skill sets for the given task which in reality will lead to failure, take more time, or resources than expected.

We would love to hear your thoughts. Leave a comment, click on the like button, and find us on Twitter @Predigt1!

#data-science#newbie#technology#womenwhocode#machine-learning
 
Share this