在 MATLAB 中调用 Python
用于无线通信的 AI 直接从 MATLAB® 中调用 Python® 库功能
与使用其他深度学习框架的同事协作，通过直接从 MATLAB 中调用 Python 来训练和测试 PyTorch®、TensorFlow™ 或 ONNX™ 模型。您还可以导入和导出函数。
工作流步骤包括数据生成、数据准备、深度神经模型训练、模型压缩、模型测试和模型部署。
这些示例重点关注训练、测试和部署工作流步骤，以演示如何在 5G 无线通信系统中使用人工智能 (AI) 来运行用于信道状态信息 (CSI) 反馈压缩和 CSI 预测技术的 PyTorch 模型。
主题
简介
- Call Python from MATLAB for Wireless
AI for wireless workflows calling Python from MATLAB to run PyTorch or TensorFlow models. (自 R2025a 起) - PyTorch Wrapper Template
You can use your own PyTorch models in MATLAB by using the Python interface.
模型训练
- Train PyTorch Channel Prediction Models
Train a PyTorch neural network for channel prediction by using data generated in MATLAB. (自 R2025a 起) - Train PyTorch Channel Prediction Models with Online Training
Enable real‐time adaptation to time‐varying wireless channels by generating each training batch in MATLAB on-the-fly to train a PyTorch GRU channel prediction network online. (自 R2026a 起) - Offline Training and Testing of PyTorch Model for CSI Feedback Compression
Train an autoencoder-based PyTorch neural network offline and test for CSI compression. (自 R2025a 起) - Online Training and Testing of PyTorch Model for CSI Feedback Compression
Train an autoencoder-based PyTorch neural network online and test for CSI compression. (自 R2025a 起)
模型测试
- Test AI-based CSI Compression Techniques for Enhanced PDSCH Throughput
Measure physical downlink shared channel (PDSCH) throughput in a 5G New Radio (NR) system, with a primary focus on AI-based compression methods for CSI feedback. (自 R2026a 起) - Apply Transfer Learning on PyTorch Model to Identify 5G and LTE Signals
Coexecution with Python to identify 5G NR and LTE signals by using the transfer learning technique on a pre-trained PyTorch™ semantic segmentation network for spectrum sensing. (自 R2025a 起) - Verify Performance of 6G AI-Native Receiver Using MATLAB and PyTorch Coexecution
Integrate a trained PyTorch network with MATLAB-based data generation to simulate an AI-native air interface. (自 R2025a 起)
模型部署
- Import TensorFlow Channel Feedback Compression Network and Deploy to GPU
Generate GPU specific C++ code for a pretrained TensorFlow channel state feedback autoencoder. (自 R2023b 起)
