Hands-On ROS for Robotics Programming

The ML pipeline

Problem-solving is split into two parts. The first is training the model according to the pipeline shown in this diagram:

Since we are assuming a simple model where the output depends linearly on the values of the features, the goal of training consists of determining the weights to be applied to each of them to obtain the prediction. Let's explain it with this mathematical formulation:

Price = W1 * area + W2 * nº rooms + W3 * distance

As you may infer, the weights, W1, W2, and W3, are the coefficients that multiply each feature. After making the sum of the three products, we obtain the predicted price. So, the training phase ...