Can twelve sold homes price the next three?
The point of a small neural net in a spreadsheet is not magic, it is the test score: the network holds back rows it never trains on and grades itself where it cannot cheat.
Home Starter Neural Net free
After you install, this is the model to open.
Teach a Tiny Neural Net House Prices
- In your spreadsheet, click the Sortia icon in the strip of icons down the right-hand edge. No strip? Click the arrow at the bottom-right to open it. You can also use Extensions, then Sortia, then Open Sortia.
- Click Start from a template and put that name in the search box.
- Pick the card with that name and click Load this template. It arrives on a new tab with real numbers already in it.
The answer
Scores first, predictions second, always in that order:
- Train R²
- ~0.999
- Test R²
- ~0.99 the honest one
- Predictions
- ~$223K / $310K / $405K
Trust the test number only, it is measured on rows the network never saw. When train is high and test collapses, the net memorised instead of learned: use fewer neurons or more rows. Training is randomized, so runs vary slightly. Any predict-a-number-from-columns question fits this shape: prices from features, churn from usage, yield from conditions.
The model
Twelve sold houses on the left (square footage and rooms in, price out) and three houses you are curious about on the right. A network with 8 hidden neurons trains for 400 passes, then predicts.
| Training data | 12 sold homes, $182K to $520K |
| Network | 8 hidden neurons, 400 epochs |
| Test score (held-out rows) | calculated |
| Three predictions | calculated |
Once it is in your sheet
- The model arrives with real numbers in it and runs as it stands, so you can press the button first and understand it second.
- Change the numbers to yours. The sheet marks which cells are inputs and which hold formulas, and most labels carry a note explaining the row.
- Press the run button at the bottom of the panel. It is labeled for the tool you are in, and the result lands on its own tab, with a written reading of it beside the figures.
Never used Google Sheets? Start here goes the whole way, in seven steps, and assumes nothing.
Next question
- How many boards do you actually have to buy?The Fewest Boards for the Cut List
- Your renovation quote is $30,000. What will it really cost?What Will the Renovation Really Cost?
- You changed two things and it worked. Which one did it?Oven, Recipe, or Both?
- Can you spot the bad batch before inspection does?Predict the Defect Before It Ships
- Is that vibration random noise or a real rhythm?Find the Hidden Vibration Frequencies
- What happens to the reserve when costs rise together?Will the Contingency Last?
Every model like this one, and the method behind them: Machine learning in Google Sheets.