Machine Learning RoadMap Part - II #62

 Performance analysis

Model evaluation helps best model that represents our data and how well the choose model in the future.

 

Resources

·         Medium blogs

·         Andrew Ng ML courses

·         Fast.ai

·         Machine learning mastery

·         Check our 100daysofmicode

 

Concept

·         Confusion matrix, accuracy, f1 score, AUC ROC, Bias variance tradeoff. R^2. MSE, Error rate etc.

 

Machine Learning

To star machine learning, first understanding the terminologies around machine learning and its types (supervised learning, unsupervised learning, Dimensionality reduction techniques, time series etc.

 

Resources

·         Machine learning by Andrew Ng (courses) fast.ai machine learning by jhon Hopkins (courser)

·         Midcourse.ai

·         Hands-on machine learning with scikit- learn and tensor flow

 

Concept

·         Different types of learning. Parameter types. Understand ML taxonomy, different types of algorithm.

 

Hyper parameter tuning

There are not model parameters and cannot trained directly from the data it is a process of choosing a set of optimal parameters to control the learning process.

 

Resources

·         Medium blogs

·         Andrew Ng ML courses

·         Fast.ai

·         Machinelearningmastery

·         Check our 100daysofmicode

 

Concepts

·         L1 norm, L2 norm, early stopping sparse regularization elastic net, mean-constrained, Graph based similarly etc.

 

Practice and Practice

Get Good hands-on experience by practicing what you learned by doing projects. Participate in competition meet follow data scientists and learn from theme.