Use chronological validation with a sample-count gap
For evaluation on later observations, TimeSeriesSplit returns split indices for observations already ordered chronologically, using earlier observations for training and the following fold for testing. Its gap parameter, which defaults to zero, excludes a specified number of samples from the end of each training set before the test set. The gap specifies a sample count, rather than an elapsed-time duration.
Conditions and limits:
- The input observations must be ordered chronologically for the split indices to represent earlier training and later testing.
- The documented interpretation of comparable fold metrics through equal test durations assumes equally spaced observations.
- A fixed number of excluded samples need not correspond to a fixed elapsed-time interval when observations are irregularly spaced.
- Chronological splitting does not jointly enforce unseen-group separation; TimeSeriesSplit ignores its groups argument.
- Chronological order and a gap do not prove that all temporal, feature or label leakage is absent.
- The supplied evidence establishes neither an optimal gap size nor improved accuracy or universal superiority over other validation designs.
Sources:
0
1
Tags
Machine Learning
Deep Learning
Supervised Learning
Dive into Deep Learning @ D2L
Data Science
Machine Learning Strategy
Machine Learning Yearning @ DeepLearning.AI
Related
When Development and Test Sets Reflect Different Populations
How Model Capacity Changes the Risk of Mixing Data Sources
One Predictor Can Work Across Multiple Data Sources
Choose evaluation data to match the real-world target
Mismatched Auxiliary Data Source
Building Dev and Test Sets Before Real Users Exist
Refreshing Evaluation Sets After a Product Launch
Using Public Web Images When No Better Future-Like Data Exists
Judging How Much to Invest in Dev and Test Sets
What should determine dev and test set selection?
True or False: You can assume the training set and test set always come from the same distribution.
Development and test sets should reflect the conditions you expect after deployment, not only the _____ available in your training pool.
Why can a simple random test split be a poor choice when the data you expect in the future is different from the data you have now?
You can usually assume the data used for training and the data used for testing come from the same distribution.
Design Dev and Test Sets for the Future
Match each concept about development and test sets to its description.
Order the steps for choosing development and test sets when future data differs from training data.
What should dev and test examples be designed to resemble?
A validation and test set must exactly match the training distribution in every project.
How should a test set be chosen when deployment data will differ?
Match each data scenario to the best dev/test set choice.
Order the reasoning steps for deciding whether a dev/test split is appropriate.
Why a Random 30 Percent Split Can Be Misleading When Future Data Will Differ
Dev and Test Splits for a Field-Photo Classifier
How to Choose Dev and Test Data When Future Data Will Differ
When to Exclude Unhelpful Training Data
Hold out entire groups to evaluate generalization to unseen groups
Use chronological validation with a sample-count gap