分享
003-Experiments with Harsh Mixing
输入“/”快速插入
003-Experiments with Harsh Mixing
1.
Setting
•
Normalize data to (0-0.25), Gaussian blur, to simulate spectrum addition
•
channel to a single value, which is then followed by a linear layer.
•
Logistic regression, BCE loss.
2.
Experiment 1: EMNIST.Digits
2.1 Config
•
ENIST digits 0-7 training
•
Random r/2 + 0.1 on the clean and all mixting
(Note this is different from Exp2, where the weights of two signals added to 1.0
)
•
White: Gaussian noise
•
Square: corrupted by small squares
•
Mix: A clean mixed with 1-3 images that are all in the target set
•
Scale: one corruption image * scale 1-3
•
Overlap: 1-3 corruption images overlapped
2.2 Results