Explore MATLAB source code curated for "训练" with clean implementations, documentation, and examples.
A MATLAB-based program that trains BP neural networks using Particle Swarm Optimization (PSO) algorithm, featuring simple and practical code implementation with enhanced training efficiency.
Image Compression with Artificial Neural Networks - MATLAB Implementation. This codebase, developed in MATLAB 6.5, provides a complete Windows-compatible solution for neural network-based image compression. The package includes all necessary files with Train.m handling network training through backpropagation algorithms and Codec.m performing image compression/decompression using the trained network weights.
"BP.m" contains the complete source code for the BP neural network model; "train.fig" shows the final training visualization; "population_data_original.fig" displays prediction result graphs; "matlab_command_window_output.txt" captures console messages during execution; "generated_data.mat" stores post-execution data files. .bmp files serve the same purpose as .fig files.
MATLAB-based face recognition using Haar-like feature training, providing a practical implementation with algorithmic insights for developers.
A novel neural network training program developed in MATLAB that enables customizable network training with enhanced algorithmic features
Character and digit training based on Radial Basis Function (RBF) neural networks, primarily implemented for license plate recognition systems with excellent performance results
This MATLAB program focuses on digital speech recognition training and identification. Due to the large size of the complete dataset, only a small sample is uploaded here. Users can create additional data using software like COOLEDIT by following these specifications: WAV files must have 8000 Hz sampling rate, mono channel, 16-bit sampling precision with Motorola PCM format. Corresponding LAB files should contain speech segment boundaries (start/end points) and phonetic content labels for training data annotation.
Comprehensive guide on using LIBSVM (a Support Vector Machine program created by Taiwan University professors), covering data preprocessing, data import, model training, and prediction with code implementation examples
A MATLAB implementation utilizing the AdaBoost algorithm for face classification, with the JAFFE facial database serving as the test dataset. The system identifies input faces by calculating similarity scores between the input image and all prototype faces in the database, then ranking the results to determine identity. Key implementation steps include: 1. Model Training: Click the [Train] button to build the recognition model using 15 face images per subject from JAFFE database 2. Test Image Selection: Click [Open] to select a face image from JAFFE for testing 3. Recognition: Click [Identify] to automatically classify the input face and display results
A wavelet neural network program designed for training and generalization applications, featuring robust algorithm implementation and performance optimization.