|
Algorithm 3 Data Screening Algorithm On the Node |
-
1:
Given from the server, i = 1 …C, where C is the number of categories.
-
Input:
A new target data:
-
2:
Obtain after feature extraction.
-
3:
Perform inference.
-
4:
if Its confidence level is greater than then
-
5:
for each do
-
5:
Calculate the distance between each and the of the corresponding pseudo-label category.
-
5:
Update the average distance once.
-
6:
if Equation (9) is satisfied then
-
7:
Output: upload .
-
8:
end if
-
9:
end for
-
10:
end if
|