[英]Which one is which ? (True Positive, True Negative, False Positive, False Negative)
I am running confusion matrix on my own custom model using Tensorflow Object Detection API. I am using Faster R-CNN Inception v2 pets.我正在使用 Tensorflow Object 检测 API 在我自己的自定义 model 上运行混淆矩阵。我正在使用 Faster R-CNN Inception v2 宠物。 I get this output:
我得到这个 output:
Processed 100 images
Processed 200 images
Processed 300 images
Processed 400 images
Processed 500 images
Processed 500 images
Confusion Matrix:
[[1281. 233.]
[ 581. 0.]]
category precision_@0.5IOU recall_@0.5IOU
0 person 0.68797 0.846103
From this matrix:从这个矩阵:
[[1281. 233.]
[ 581. 0.]]
Which one is true positive, true negative, false positive, false negative?哪一个是真阳性、真阴性、假阳性、假阴性?
I am using code from this github .我正在使用来自这个github的代码。 It said that this link would provide more explanation about this code, but the post went missing.
它说此链接将提供有关此代码的更多解释,但该帖子丢失了。 So, i am confused.
所以,我很困惑。
Also, can i calculate accuracy from this results?另外,我可以根据这个结果计算准确度吗? Sorry if i'm wrong.
对不起,如果我错了。
Please check below image.请检查下图。
More information about confusion matrix can be found here.有关混淆矩阵的更多信息,请参见此处。 https://www.analyticsvidhya.com/blog/2020/04/confusion-matrix-machine-learning/
https://www.analyticsvidhya.com/blog/2020/04/confusion-matrix-machine-learning/
True positive: 1281 True negative: 0. False Negative: 581. False Positive: 233.真阳性:1281 真阴性:0。假阴性:581。假阳性:233。
Confussion matrix is a performance measurement for machine learning classification problem where output can be two or more classes or simplify we can assume CM calculate accuracy/loss your model.混淆矩阵是机器学习分类问题的性能度量,其中 output 可以是两个或多个类或简化我们可以假设 CM 计算准确度/损失您的 model。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.