Table 2 Summary table of the model notations with interpretation.
From: CAD-PsorNet: deep transfer learning for computer-assisted diagnosis of skin psoriasis
Notations | Interpretation |
|---|---|
DK | Kernel size of the layer |
DF | Feature map size |
\(\:M\) | Number of input channels |
\(\:N\) | Number of output channels |
α | Width multiplier to control the number of channels in depth wise convolutional layer |
ρ | Resolution multiplier to control resolution of input images |
n | Number of observations |
\(\:\beta\:\) | Moving average parameter |
\(\:{m}_{t-1}\) | Aggregate of gradient at t-1 |
\(\:\gamma\:\) | Learning rate |
\(\:\eta\:\) | Initial global learning rate |
\(\:\in\:\) | small constant added to eliminate the problem of division by zero |
\(\:{w}_{t}\) | Parameter value a time t |
\(\:{G}_{ii}\) | G represent a diagonal matrix where each element Gi, i corresponds to the sum of the squares of the historical gradients for the i-th parameter |
