AI 900 - Machine Learning Fundamentals

Machine Learning Fundamentals - 19 Cards
Click here to toggle all cards
ML Process Result
A model which can predict
Inputs
Features (What to use? - area, location,size ..) and Label (What to predict? - house price)
Supervised Learning
Clearly Defined Ouput - Features and Labels
Regression
Label is a numeric value with a range of possibilities - How much will it rain tomorrow?
Classification
Label has limited set of possibilities - Will it rain today?
Unsupervised Learning
No Label
Clustering
Divide customers into groups - Group similar entities based on their features
Steps in Machine Learning Projects
Obtain data, clean data, feature engineering, create model, evaluate accuracy, and deploy the model.
Training
Creating a model
Evaluation
Checking if the model works
Inference
Using the model for predictions
Features
Inputs in a dataset
Label
Output/Prediction in a dataset
Training Dataset
Dataset used to create a model
Validation Dataset
Dataset used to validate the model and choose the right algorithm
Testing Dataset
Dataset used for final testing before deployment
Azure Machine Learning
Simplifies creation of your models
Automated machine learning
Build custom models with minimum ML expertise
Azure Machine Learning designer
Build Your Own Models with Data Scientists