Hornbill.AI
- Mar 4
- 2 min read

Building Your First Predictive Model On Hornbill.AI

Hornbill.AI is helping hundreds of firms train, deploy & host state-of-the-art machine learning models without the overheads — both in terms of time & resources traditionally required. Here’s a quick introduction to the platform, and simple steps to deploy your very own neural network to predict real estate prices in Boston.

Step 1: Log in to your Hornbill.AI account

No frills here, just enter your email & password; or register if you haven’t already.

Step 2: Download the Boston housing dataset

The Boston housing dataset is widely used across the ML community for benchmarking regression tasks; you can download the .csv file here (go to the link & just press ctrl+S/command+S; credits: runnily @ github)

The dataset contains multiple columns, also referred to as features, regarding regarding elements of properties across Boston: from CRIM (crime rates in the town) to NOX (nitric oxide concentration), I’ve put down a brief for the variables here.

Variables in order: CRIM Per capita crime rate by town ZN Proportion of residential land zoned for lots >25,000 sq.ft. INDUS Proportion of non-retail business acres per town CHAS Charles River dummy variable NOX Nitric oxides concentration (parts per 10 million) RM Average number of rooms per dwelling AGE Proportion of owner-occupied units built prior to 1940 DIS Weighted distances to five Boston employment centres RAD Index of accessibility to radial highways TAX Full-value property-tax rate per $10,000 PTRATIO pupil-teacher ratio by town B 1000(Bk - 0.63)^2 where Bk is the proportion of blacks by town LSTAT % lower status of the population MEDV Median value of owner-occupied homes in $1000's

Variables in order: CRIM Per capita crime rate by town ZN Proportion of residential land zoned for lots >25,000 sq.ft. INDUS Proportion of non-retail business acres per town CHAS Charles River dummy variable NOX Nitric oxides concentration (parts per 10 million) RM Average number of rooms per dwelling AGE Proportion of owner-occupied units built prior to 1940 DIS Weighted distances to five Boston employment centres RAD Index of accessibility to radial highways TAX Full-value property-tax rate per $10,000 PTRATIO pupil-teacher ratio by town B 1000(Bk - 0.63)^2 where Bk is the proportion of blacks by town LSTAT % lower status of the population MEDV Median value of owner-occupied homes in $1000's

We’ll try to predict MEDV (median value in $1000s) of the house given the other features.

Step 3: Upload the dataset to Hornbill.AI

Uploading a dataset to Hornbill.AI is a pretty painless process, just go to your “Datasets” tab → click on the “Upload” button → enter some context about the dataset & press upload, that’s it!

Step 3: Deploy your project

Awesome, now you’ve deployed your dataset & are almost there! Go to your “Projects” page → click on “Deploy” → choose the dataset as boston-housing → choose “medv” (median value of the house) as the target feature → press “Deploy”

Once you’ve pressed “Deploy”, you can see your models training in real-time and your model should be computed in under a minute.

Step 4: Start forecasting

Cool! You’ve trained, deployed & hosted your very own ML model in just minutes. Now you can press “interact” to view more details about your project. In your project, go to the “Forecasts” tab, input some data & get the predicted price of that house.

That’s it, you can now integrate your model in a host of different way: from shareable links to API connections — the possibilities with Hornbill.AI are limitless!