Example 1 Data

 

TensorFlow Data Collection

The data used in Example 1, is a list of car objects like this:

{
  "Name": "chevrolet chevelle malibu",
  "Miles_per_Gallon": 18,
  "Cylinders": 8,
  "Displacement": 307,
  "Horsepower": 130,
  "Weight_in_lbs": 3504,
  "Acceleration": 12,
  "Year": "1970-01-01",
  "Origin": "USA"
},
{
  "Name": "buick skylark 320",
  "Miles_per_Gallon": 15,
  "Cylinders": 8,
  "Displacement": 350,
  "Horsepower": 165,
  "Weight_in_lbs": 3693,
  "Acceleration": 11.5,
  "Year": "1970-01-01",
  "Origin": "USA"
},

The dataset is a JSON file stored at:

https://storage.googleapis.com/tfjs-tutorials/carsData.json

---

Cleaning Data

When preparing for machine learning, it is always important to:

• Remove the data you don't need
• Clean the data from errors

---

Remove Data

A smart way to remove unnecessary data, it to extract only the data you need.

This can be done by iterating (looping over) your data with a map function.

The function below takes an object and returns only x and y from the object's Horsepower and Miles_per_Gallon properties:

function extractData(obj) {
  return {x:obj.Horsepower, y:obj.Miles_per_Gallon};
}