Machine Learning 101

Machine Learning: The fundamental idea of machine learning is to use data from past observations to predict unknown outcomes or values.

ML Model: A machine learning model is a software application that encapsulates a?function?to calculate an output value based on input values.?

Model Training: The process of defining function is known as?Model Training.

Inferencing: When we use this function to predict new values based on our new input, it’s called?inferencing

Feature: To train the model, we provide our past observation data, providing its attribute of the thing being observed (example for animal, it’s no of legs[x1], height[x2], length[x3] etc).

Label: ??This is known value of the thing we want to train a model to predict?(example: based on legs[x1], height[x2] , length[x3] etc we want our model to predict is it mammal[y2] or reptile[y2] )

Algorithm: ?Applied to data, to determine a relationship between the features [x1, x2, x3] and the label[y], to generalize that relationship as a calculation that can be performed on?x?to calculate?y. basically try to?fit?a function to the data, in which the values of the features can be used to calculate the label.

Vector: An observation consists of multiple feature values, so?x?is actually an array with multiple values), like,?[x1,x2,x3.], it’s called vector and hence we need vector database to store it.

The result of the algorithm is a?model?that encapsulates the calculation derived by the algorithm as a?function(f)

So, training the model in mathematical notation is nothing but y = f(x)

Once the training?phase is complete, the trained model can be used for?inferencing.

And the model is essentially a software program that encapsulates the function produced by the training process.?

?Machine Learning Type

Supervised ML: Algorithm in which the training data includes both?feature?values and known?label?values. Supervised machine learning is used to train models by determining a relationship between the features and labels in past observations, so that unknown labels can be predicted for features in future cases.

Further divided into Regression (numeric value) & classification(categorization)

Unsupervised ML: It involves training models using data that consists only of?feature?values without any known label and its algorithms determine relationships between the features of the observations in the training data.

Clustering is example of Unsupervised ML.

to be continued......