What Is Deep Learning Toolbox?
Deep Learning Toolbox™ provides functions, apps, and Simulink® blocks for designing, training, implementing, and simulating deep neural networks. The toolbox provides a framework to create and use many types of networks, such as convolutional neural networks (CNNs) and transformers. You can visualize and interpret network predictions, verify network properties, and compress neural networks with quantization, projection, and pruning.
With the Deep Network Designer app, you can design, edit, and analyze networks interactively, import pretrained models, and export networks to Simulink. The toolbox lets you interoperate with other deep learning frameworks. You can import PyTorch®, TensorFlow™, and ONNX™ models for inference, transfer learning, simulation, and deployment. You can also export models to TensorFlow and ONNX.
You can automatically generate C/C++, CUDA®, and HDL code for trained networks.
Published: 12 Apr 2024
Deep Learning Toolbox provides functions, apps, and Simulink blocks that enable engineers to use deep learning applications like visual inspection, wireless communications, reduced order modeling, and computer vision.
With the toolbox, you can test and deploy deep neural networks. Deep learning with Simulink allows you to simulate the performance of neural networks within larger systems for deployment to CPUs, GPUs, FPGAs, and systems on chip. You can automatically generate C, C++, Cuda, and HDL code for your trained networks.
The toolbox integrates with Python-based deep learning frameworks. Import PyTorch, TensorFlow, and ONNX models with one line of code. Export networks to TensorFlow and ONNX just as easily. Coexecute any Python-based model in MATLAB and Simulink. With the Deep Network Designer app, you can interactively import PyTorch and TensorFlow models and load built-in models. You can analyze and modify models for transfer learning and export them to Simulink.
Deep Learning Toolbox lets you design and train network architectures like CNNs LSTMs, GANs, and transformers. To train your networks, you can automatically label, process, and augment image, video, and signal data. In addition to simulation, you can use robustness verification and explainability techniques. This ensures that your deep learning models work as intended before deploying to target.
You can compress neural networks with quantization, projection, or pruning. These methods reduce memory use and increase inference performance for deployment to resource-constrained devices. The toolbox provides hundreds of reference examples from deep learning fundamentals to advanced applications. For more information on Deep Learning Toolbox or how to incorporate deep learning into your projects, visit the links below.