Multiple order modeling for deep learning
Introduction
This is an example of multiple order modeling for accuracy improvement in deep neural networks.
Different approaches are shown on how to use the outputs of a category prediction model as predictors for a second model.
Time series instances of samples are used as multiple inputs (for example N frames of a video is used as N image inputs) for model,
and those N number of predicted output (probability density) is used as predictors for the second model.
Data
We attach a set of simulated data for testing this approach.
Details regarding the data is available in comment section of FE_DataLoad.m
Supporting function
A function (trainClassifier.m) attached here for training data with SVM algorithm is called by the scripts.
This function is generated using MATLAB's CalssifierLearner App's code generation functionality
Zitieren als
Mohammad (2024). Multiple order modeling for deep learning (https://github.com/muquitMW/multiple_order_modeling/releases/tag/1.1), GitHub. Abgerufen .
Kompatibilität der MATLAB-Version
Plattform-Kompatibilität
Windows macOS LinuxTags
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!Live Editor erkunden
Erstellen Sie Skripte mit Code, Ausgabe und formatiertem Text in einem einzigen ausführbaren Dokument.
Version | Veröffentlicht | Versionshinweise | |
---|---|---|---|
1.1 | See release notes for this release on GitHub: https://github.com/muquitMW/multiple_order_modeling/releases/tag/1.1 |