模型预测指标的意思

参考：https://blog.goodaudience.com/introduction-to-1d-convolutional-neural-networks-in-keras-for-time-sequences-3a7ff801a2cf

Here is a brief recap of what those scores mean:

• Accuracy: The ratio between correctly predicted outcomes and the sum of all predictions. ((TP + TN) / (TP + TN + FP + FN))
• Precision: When the model predicted positive, was it right? All true positives divided by all positive predictions. (TP / (TP + FP)) : 当模型预测是真时？有大可能是真？(假的但被预测为真的)
• Recall: How many positives did the model identify out of all possible positives? True positives divided by all actual positives. (TP / (TP + FN))  正向样本中有多少被预测出
• F1-score: This is the weighted average of precision and recall. (2 x recall x precision / (recall + precision))

 

 横坐标是预测的lable: 纵坐标是真是的label:

看纵坐标Walking: 时间预测的结果是：2397中：2351预测正确，36个预测为下楼梯，10个预测为上楼梯

验证下f1-score: (2 x recall x precision / (recall + precision)):  2*0.76*0.78/(0.76+0.78) = 0.7698