1. Github download
This is used to draw the mAP curve.
https://github.com/Cartucho/mAP
This is used to obtain the txt required to draw the mAP curve
https://github.com/bubbliiiiing/count-mAP-txt
2. The concept of TP TN FP FN
There are a total of 4 letters in TP TN FP FN, which are TFPN.
T is True;
F is False;
P is Positive;
N is Negative.
T or F represents whether the sample is correctly classified.
P or N represents whether the sample was originally a positive sample or a negative sample.
TP (True Positives) means that it has been classified as a positive sample, and it has been classified correctly.
TN (True Negatives) means that it was classified as a negative sample, and it was classified correctly.
FP (False Positives) means that it was classified as a positive sample, but it was classified incorrectly (in fact, this sample is a negative sample).
FN (False Negatives) means that it was classified as a negative sample, but it was wrongly classified (in fact, this sample is like this).
In the process of mAP calculation, the three concepts of TP, FP and FN are mainly used.
3. precision and recall
TP is an example that the classifier thinks is a positive sample and is indeed a positive sample. FP is an example that the classifier thinks is a positive sample but is actually not a positive sample. Precision is translated into Chinese as "the part that the classifier thinks is a positive class and is indeed a positive class accounts for all categories." The device considers it to be the proportion of the positive kind."
TP is an example that the classifier thinks is a positive sample and is indeed a positive sample. FN is an example that the classifier thinks is a negative sample but is actually not a negative sample. Recall is translated into Chinese as "the part that the classifier thinks is a positive class and is indeed a positive class accounts for all It’s definitely a proportion of the right kind.”