forked from animatedread/Warrior_EA
afee4cc4c1
Reduce weight decay from 0.01 to 0.001 across all backends (Network.cl, Network.mqh, WarriorCPU.cpp) to fix a training collapse issue. The original 0.01 AdamW default caused discriminative weights to decay below the calibration-capped class-prior offsets, resulting in a monotonically shrinking per-bar logit spread and eventual constant Neutral predictions (argmax degenerated once evidence tilt dropped under the prior tilt). The new value 0.001 lifts the evidence ceiling 10× while still bounding long-run weight growth, restoring effective discrimination. Note: this change must remain in sync across all four backends. |
||
|---|---|---|
| .. | ||
| ArrayLayer.mqh | ||
| BufferDouble.mqh | ||
| LayerDescription.mqh | ||
| Network.cl | ||
| Network.mqh | ||
| NeuronCPU.mqh | ||
| NeuronDirectML.mqh | ||
| NeuronOCLConvPool.mqh | ||
| NeuronPrimitives.mqh | ||