7 questions found
What is Training vs Testing Data
Beginner
Training data is the data used to teach a machine learning model, while testing data is a separate set used to check how well the model performs on examples it has never seen before.
Real-world example
A bank builds a loan approval model on data from the last five years and tests it on the most recent six months of applications to see how it would have performed in practice.
Common follow-ups: What is a validation set, What is cross validation, What is overfitting
Machine Learning Fundamentals
Deep Learning
Supervised & Unsupervised Learning
Why is Training vs Testing Data important in Machine Learning Fundamentals
Beginner
Separating training and testing data is important because it shows whether a model actually learned useful patterns or just memorized the training examples. Without this split you cannot trust the reported accuracy.
Real-world example
A bank builds a loan approval model on data from the last five years and tests it on the most recent six months of applications to see how it would have performed in practice.
Common follow-ups: What is a validation set, What is cross validation, What is overfitting
Machine Learning Fundamentals
Deep Learning
Supervised & Unsupervised Learning
How does Training vs Testing Data work
Beginner
A dataset is usually split into a training portion, often seventy to eighty percent, and a testing portion, the remaining twenty to thirty percent. The model only sees the training data during learning and is evaluated afterward on the testing data.
Real-world example
A bank builds a loan approval model on data from the last five years and tests it on the most recent six months of applications to see how it would have performed in practice.
Common follow-ups: What is a validation set, What is cross validation, What is overfitting
Machine Learning Fundamentals
Deep Learning
Supervised & Unsupervised Learning
What are the key parts or types of Training vs Testing Data
Intermediate
The main parts of this process are the training set, the validation set used for tuning, and the final test set used only once at the end to report performance.
Real-world example
A bank builds a loan approval model on data from the last five years and tests it on the most recent six months of applications to see how it would have performed in practice.
Common follow-ups: What is a validation set, What is cross validation, What is overfitting
Machine Learning Fundamentals
Deep Learning
Supervised & Unsupervised Learning
What are common mistakes to avoid with Training vs Testing Data
Intermediate
A common mistake is letting information from the test set leak into training, for example by scaling the entire dataset together before splitting, which gives falsely optimistic results.
Real-world example
A bank builds a loan approval model on data from the last five years and tests it on the most recent six months of applications to see how it would have performed in practice.
Common follow-ups: What is a validation set, What is cross validation, What is overfitting
Machine Learning Fundamentals
Deep Learning
Supervised & Unsupervised Learning
What is a real world example of Training vs Testing Data
Advanced
A bank builds a loan approval model on data from the last five years and tests it on the most recent six months of applications to see how it would have performed in practice.
Real-world example
A bank builds a loan approval model on data from the last five years and tests it on the most recent six months of applications to see how it would have performed in practice.
Common follow-ups: What is a validation set, What is cross validation, What is overfitting
Machine Learning Fundamentals
Deep Learning
Supervised & Unsupervised Learning
What are best practices for Training vs Testing Data
Advanced
Always split data before doing any preprocessing, use random or time based splits depending on the problem, and never look at the test set until the model is fully finished.
Real-world example
A bank builds a loan approval model on data from the last five years and tests it on the most recent six months of applications to see how it would have performed in practice.
Common follow-ups: What is a validation set, What is cross validation, What is overfitting
Machine Learning Fundamentals
Deep Learning
Supervised & Unsupervised Learning