Data for: Redshift Prediction with Images for Cosmology using a Bayesian Convolutional Neural Network with Conformal Predictions

These files contain the predictions from the CNN and BCNN model from the paper titled: "Redshift Prediction with Images for Cosmology using a Bayesian Convolutional Neural Network with Conformal Predictions" (Jones et al. 2024). These files will allow reproduction of the performance metrics described in the paper. full_prediction_set_CNN.csv - predictions for the redshift using the CNN model of the entire datasetcnn_evaluation.csv - predictions from just the evaluation dataset that was not used in training Columns are: photoz - predicted photoz from the modelspecz - spectroscopic redshiftobjectid - object ID from HSC PDR2 data release (Aihara et al. 2019) full_prediction_set_BCNN.csv - predictions for the redshift using the BCNN model of the entire datasetbcnn_evaluation.csv - predictions from just the evaluation dataset that was not used in training Columns are: photoz - predicted photoz from the modelspecz - spectroscopic redshiftobjectid - object ID from HSC PDR2 data release (Aihara et al. 2019)photoz_uncertainty - uncertainty in the predicted photoz本数据集包含论文《基于共形预测贝叶斯卷积神经网络的宇宙学图像红移预测》（Jones等，2024）中卷积神经网络（Convolutional Neural Network, CNN）与贝叶斯卷积神经网络（Bayesian Convolutional Neural Network, BCNN）的预测结果，可用于复现论文中所述的各项性能指标。 `full_prediction_set_CNN.csv`：基于CNN模型对全量数据集生成的红移预测结果；`cnn_evaluation.csv`：仅使用训练阶段未使用的评估数据集得到的预测结果。 两文件的列字段统一为： - photoz：模型预测的测光红移（photometric redshift，简称photoz） - specz：光谱红移（spectroscopic redshift，简称specz） - objectid：来自HSC PDR2数据发布的天体ID（Aihara等，2019） `full_prediction_set_BCNN.csv`：基于BCNN模型对全量数据集生成的红移预测结果；`bcnn_evaluation.csv`：仅使用训练阶段未使用的评估数据集得到的预测结果。 其列字段包含： - photoz：模型预测的测光红移 - specz：光谱红移 - objectid：来自HSC PDR2数据发布的天体ID（Aihara等，2019） - photoz_uncertainty：预测测光红移的不确定性