GPU Coder為深度學習、嵌入式視覺、自主系統從MATLAB產生經過優化的CUDA程式碼。來看看怎麼利用GPU Coder自動產生可在你的電腦或NVIDIA GPU執行的GPU程式碼。在這段演講，我們將探索以下幾個重點：
- 簡單介紹MATLAB、利用MATLAB進行平行運算、從MATLAB產生程式碼
- 為深度學習、嵌入式視覺、自主系統從MATLAB程式碼產生經過優化的CUDA程式碼
- 將產生出來的程式碼整合為原始碼、靜態或動態函式庫
- 產生出來的CUDA可攜於NVDIA GPUs
- 產生出的程式碼可呼叫經過優化的NVIDIA CUDA函式庫，包含cuDNN、cuSolver、cuFFT、cuBLAS
- MATLAB演算法和產生出的程式碼與手寫CUDA程式碼的整合
- 在NVIDIA Tesla®、NVIDIA Tegra®、Jetson TX2等GPUs原型化演算法
- 在MATLAB使用產生的CUDA程式碼，以加速需要密集運算的MATLAB程式碼(mex)
GPU Coder generates optimized CUDA code from MATLAB code for deep learning, embedded vision, and autonomous systems. See how you can use GPU Coder to automatically generate and execute GPU Code on your desktop or embedded NVIDIA GPU. In this presentation, we will explore the following:
- Brief Introduction to MATLAB, parallel computing using MATLAB, and code generation from MATLAB.
- Generate optimized CUDA code from MATLAB code for deep learning, embedded vision, and autonomous systems
- Integrate generated code as source code and static or dynamic libraries
- Generated CUDA is portable across NVIDIA GPUs
- Generated code calls optimized NVIDIA CUDA libraries, including cuDNN, cuSolver, cuFFT, and cuBLAS
- Incorporate handwritten CUDA code into MATLAB algorithms and generated code
- Prototype algorithms on GPUs such as the NVIDIA Tesla®, NVIDIA Tegra®, and the Jetson TX2.
- Use generated CUDA code within MATLAB to accelerate computationally intensive portions of your MATLAB code(mex)
|
