l2181265的博客

混淆矩阵(Confusion Matrix)label poslabel negpred posTPFPpred negFNTNTP(真正例)：预测+ 实际+FP(假正例)：预测+ 实际-TN(真负例)：预测- 实际-FN(假负例)：预测- 实际+Precision(精确率，P)P=TPTP+FPP= \frac {TP} {TP+FP}P=TP+FPTP​TPR(真正例率，灵敏度，Sensitivity，召回率，Recall)TPR=TPT

定义网络：import torch.nn as nnimport torch.nn.functional as Fclass Net(nn.Module): def __init__(self): super(Net, self).__init__() self.conv1 = nn.Conv2d(3, 6, 5) self.con...