fastmachinelearning/wa-hls4ml-projects
收藏Hugging Face2025-07-23 更新2025-11-01 收录
下载链接：
https://hf-mirror.com/datasets/fastmachinelearning/wa-hls4ml-projects
下载链接
链接失效反馈官方服务：
资源简介：
wa-hls4ml数据集是一个关于神经网络转换为HLS代码的项目数据集，包含了完整的日志、HLS代码、VHDL代码、中间表示和源Keras模型。数据集分为两部分：合成神经网络和真实科学应用模型。合成部分用于探索FPGA资源和延迟，而真实应用模型用于测试模型对新配置和架构的泛化能力。数据集由Fast Machine Learning Lab创建，并由AMD Xilinx Alveo U250 FPGA板合成。
The wa-hls4ml dataset is a project dataset about converting neural networks into HLS code, including complete logs, HLS code, VHDL code, intermediate representations, and source Keras models. The dataset is divided into two parts: synthetic neural networks and realistic scientific application models. The synthetic part is used to explore FPGA resources and latency, while the realistic application models are used to test the generalization ability of the model to new configurations and architectures. The dataset was created by Fast Machine Learning Lab and synthesized on the AMD Xilinx Alveo U250 FPGA board.
提供机构：
fastmachinelearning
