ML Libraries

Learning Objectives Define what an ML library is and explain its purpose. Identify at least three popular ML libraries (e.g., Scikit-learn, TensorFlow, PyTorch). Describe the core components of an ML workflow: data loading, model creation, training, and prediction. Differentiate between a classification and a regression problem. Read and interpret simple code that uses an ML library to train a model and make a prediction. Explain the importance of splitting data into training and testing sets. Ever wonder how your phone's camera can recognize faces or how Netflix knows exactly what movie you want to watch next? 🤖 It's not magic; it's the power of ML libraries! In this lesson, we'll explore Machine Learning (ML) libraries, which are powerful toolkits tha...

TermDefinitionExample ML LibraryA collection of pre-written code, functions, and classes that helps developers build and deploy machine learning models quickly and efficiently.Scikit-learn is a popular ML library in Python. Instead of writing a complex sorting algorithm yourself, you just call a function; similarly, instead of coding a decision tree from scratch, you can use one from Scikit-learn. ModelIn ML, a model is an object that has been 'trained' on a dataset to recognize certain patterns. It's the 'brain' of the operation that makes predictions or decisions.A model could be trained on thousands of pictures of cats and dogs. Once trained, it can look at a new picture and predict whether it contains a cat or a dog. DatasetA collection of data, typically orga...

The Standard ML Workflow Pattern 1. Import Library -> 2. Load Data -> 3. Create Model Object -> 4. Train Model -> 5. Make Prediction This five-step sequence is the fundamental pattern for using most ML libraries. It provides a structured way to go from raw data to a useful prediction. Model Instantiation Syntax model = Library.ModelType(parameters) Similar to creating an object from a class in OOP, you create a model 'object' from a specific model 'class' provided by the library. You can often customize the model by passing arguments (parameters). Training Syntax: The `.fit()` Method model.fit(training_features, training_labels) The `.fit()` method is the standard command used in many libraries to start the training process. It takes t...