Table 1 Hyperparameters for Fold Classification Architecture
From: DeepFrag-k: a fragment-based deep learning approach for protein fold recognition
Layer | Layer type | # of Units | Unit Type | Size | Stride |
|---|---|---|---|---|---|
Input | # of fragments | ||||
COV_1 | Convolution | 10 | ReLU | 1,10 | 1,1 |
MP_1 | Max Pool | 1,10 | 1,1 | ||
ST | Stacking | ||||
COV_2 | Convolution | 100 | ReLU | 10,10 | 1,1 |
MP_2 | Max Pool | 5,5 | 5,5 | ||
FC_1 | Fully Connected | 100 | ReLU | ||
FC_2 | Fully Connected | 100 | ReLU | ||
Output | SoftMax | # of folds | Logistics |