Quantification of CO2 hotspot emissions from OCO-3 SAM CO2 satellite images using deep learning methods - data and weights

Release for "Quantification of CO2 hotspot emissions from OCO-3 SAM CO2 satellite images using deep learning methods", submitted to "Atmospheric Chemistry and Physics- Train, validation and test dataset for Lippendorf, Boxberg, and Turow CNN applications.- Test dataset for OCO3 SAM application.- Weights and architecture of the trained CNN.本数据集为提交至《大气化学与物理（Atmospheric Chemistry and Physics）》的论文《基于深度学习方法量化OCO-3 SAM二氧化碳卫星影像中的CO₂热点排放》的配套发布内容，具体包括： - 用于Lippendorf、Boxberg及Turow三地卷积神经网络（Convolutional Neural Network, CNN）应用的训练、验证与测试数据集； - 用于OCO-3 SAM应用的测试数据集； - 已训练完成的卷积神经网络的权重与架构。