Large language models (LLMs) continue to help users answer their questions, generate presentations and many other assorted tasks. These models are also good at generating code for user specified tasks. In an earlier post, I had used Chat-GPT to provide code for a regression task. But Chat-GPT was unable to execute the code.
Recently, Google has given its large language model, Bard, the computational capabilities as well. With this capability, Bard not only gives you the complete code but if requested will execute for you. This is a great feature for those who are trying to learn machine learning. In this post, I will show you my interaction with Bard to build and execute a regression model using the data supplied via a URL. So continue to read on.
Interaction with Bard
To ensure that Bard will execute the code as well, I started off by asking Bard whether it could generate code for machine learning.
As we see, Bard not only gave a complete code example but also provided the output. It then went on to give another example with an explaination.
Next, I ask whether I can supply my own data.
Now is the time to supply the data and ask for execution to build the model.
As you can see, Bard has broken down the modelling code into different stages explaining what is being done at each stage. It then goes on to give additional tips for creating a regression model.
Next, I ask Bard to execute to visualize data.
So as you can see from my interaction with Bard, you can now give your own data to Bard and have it do modeling and visualization for you. Learning ML doesn’t get any easier than this. So go and interact with Bard.
From Data to Decisions 