From: A new method based on stacked auto-encoders to identify abnormal weather radar echo images
Input: train pictures after integral projection: X, train labels: L | |
Output: weight matrices: \( {\left\{\overset{\wedge }{W_1^h}\right\}}_{h=1}^{H+1} \) , biases: \( {\left\{\overset{\wedge }{W_0^h}\right\}}_{h=1}^{H+1} \)/*H: number of hidden layers*/ | |
Step 1: the training of hidden layers: | |
â‘ Initialization: Y0 = X | |
② Greedy layer-wise training (h ∈ {1, …, H}) | |
- Acquire the parameters \( \left\{{W}_1^h,{W}_0^h\right\} \) for the h-th hidden layer | |
- \( {Y}_h=f\left({W}_1^h{Y}_{h-1}+{W}_0^h\right) \) | |
Step 2: fine-tuning the whole network: | |
â‘ Initialization:\( {\left\{\overset{\wedge }{W_1^h}={W}_1^h,\overset{\wedge }{W_0^h}={W}_0^h\right\}}_{h=1}^H,\left\{{\overset{\wedge }{W}}_1^{H+1},{\overset{\wedge }{W}}_0^{H+1}\right\} \)=random | |
â‘¡ Back-propagation with a gradient-descent theory |