Created with ChatGPT

I remember spending days, if not weeks, building machine learning models from scratch, cleaning and normalizing the data, and visualizing the best models, tuning parameters by hand.

With each project, I was a craftsman.

It felt good, it was gratifying.

I loved that slowness.

It made me feel in control. Grounded. Human.

AI Came Knocking on My Doorstep?

Photo by Dima Pechurin on Unsplash

Later, I found an article on Medium about this new tool. Can everything be automated? Would that be the case?

It was great, but not very great in Machine Learning.

But I did my best.

I even gave it the information in the dataset! And it helps.

But it was not optimal. Why not simply load a dataset into it?

They Listened to Us!

Photo by Franco Antonio Giovanella on Unsplash

And OpenAI listened to us. And yes, this tool’s name is ChatGPT, and this feature is a code interpreter.

I even used plug-ins to enable this feature, as they first introduced this tool to US users.

This new feature allows users to upload data, and boom.

Reference

A few tweaks to it, and it could automate the entire ML pipeline.

No notebooks. No manual tweaks.

I laughed. And clicked. And sat. For hours.

I Didn’t Trust It — At First

Photo by Joseph Chan on Unsplash

What man in his right mind would think a complicated thing like that could be summed up in a few clicks?

I felt worthless for a while.

Wouldn’t that ruin the craft? Wouldn’t that make my skills … irrelevant?

I even remember searching for new things to learn so I wouldn't be replaced.

Reference

I always know it’s worth it. I recall debates with my peers about its value.

They told me it was a toy.

Not as smart as a washing machine, a colleague once told me, could you believe?

However, learning AI has evolved, and with ChatGPT, it has changed for good.

Reference

One Dataset. One Upload. One Surprise.

Photo by Ben White on Unsplash

It loads, prepares, and fills the data, and chooses a model to fit.

By the time I’d even had my first sip of coffee, it had trained and evaluated the model.

R² score?

The only thing better than what I’d hand-built last week.

The Inner Shift

Photo by Des Récits on Unsplash

I recall telling my boss about it. Instead, we need to incorporate AI into our work.

Otherwise, we will be left behind.

It didn’t take my job.

It gave me a new one, why?

Now I have a lot more AI-related stuff to do.

So, rather than hours spent on data projects, I can now automate them.

Reference

The craft wasn’t dead. It had just ascended a level.

I mentioned it in the context of doing the work, just like I had an assistant for free!

Well, not for free, but rather for $20 a month.

The Return

Photo by Nonsap Visuals on Unsplash

Machine learning doesn’t just belong to engineers anymore.

These days, it is for everyone.

It’s for the thinkers and strategists among us.

All you need is a spark, a taste for the thing itself.

AI will not replace you. Instead, it will help you break boundaries with ideas.

Our Contribution

Photo by Artist Istanbul on Unsplash

With this intent, we also want to contribute.

Perhaps we can also refine the process a bit.

By using AI tools, of course.

All you have to do is upload your CSV, and in a couple of seconds, the data overview and data exploration parts of your data project will be completed.

Data Overview

But let’s say you want to know which factors have gained the most years in life? Click on AI Assistant and prompt it, and voila!

Talk with your data

Or, let’s say you want to build an ML model. Select Model building and select your target feature, the model you want to develop, along with test size and the predictive features, and click on Build model.

Model Building

And voila, once again!

Evaluation Metrics

But what happened? Perhaps your variables were not numeric, and the data needs to be processed accordingly. (scaling, encoding?). Our agent did it under the hood.

Feature Type Classification

Now you have a full prediction, and you did this by just clicking. I told you machine learning does not belong to the engineers anymore. Visit our platform to discover this feature!

Final Thoughts

Don’t say sorry or hesitate anymore, this is the new era, try building your model by just clicking!

If you are using ChatGPT, simply craft the prompts, prepare your CSV, and upload it to the AI.

And voila! Now you are like us. Thanks for reading this one!

If you want to craft prompts and automate the entire data analysis process with our agents, here are our assistants.

Here are the free resources.

Here is the ChatGPT cheat sheet.

Here is the Prompt Techniques cheat sheet.

Here is my NumPy cheat sheet.

Here is the source code of the “How to be a Billionaire” data project.

Here is the source code of the “Classification Task with 6 Different Algorithms using Python” data project.

Here is the source code of the “Decision Tree in Energy Efficiency Analysis” data project.

Here is the source code of the “DataDrivenInvestor 2022 Articles Analysis” data project.

“Machine learning is the last invention that humanity will ever need to make.” Nick Bostrom