Table 2.
Parameter | Value |
---|---|
Initial learning rate | 0.001 |
Learn rate schedule | Piecewise |
Learn drop period | 100 |
Learn drop factor | 0.001 |
Gradient threshold | 2 |
Parameter | Value |
---|---|
Initial learning rate | 0.001 |
Learn rate schedule | Piecewise |
Learn drop period | 100 |
Learn drop factor | 0.001 |
Gradient threshold | 2 |