mandofin/Warrior_EA
Watch
1
0
Fork
You've already forked Warrior_EA
 0
forked from animatedread/Warrior_EA
Code Pull requests Activity
Warrior_EA/DirectML
History
Exact
Repository files (latest commit first)
Filename Latest commit message Latest commit date
AnimateDread afee4cc4c1 fix: lower WEIGHT_DECAY to 0.001 to prevent argmax degeneration to Neutral 
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.
2026-07-19 17:05:58 -04:00
..
build.bat feat: add max-pooling and convolution OpenCL kernels, clean up barrier and signal code 2026-07-13 03:23:39 -04:00
build_cpu.bat feat: add max-pooling and convolution OpenCL kernels, clean up barrier and signal code 2026-07-13 03:23:39 -04:00
WarriorCPU.cpp fix: lower WEIGHT_DECAY to 0.001 to prevent argmax degeneration to Neutral 2026-07-19 17:05:58 -04:00
WarriorCPU.dll fix: lower WEIGHT_DECAY to 0.001 to prevent argmax degeneration to Neutral 2026-07-19 17:05:58 -04:00
WarriorCPU.exp fix: lower WEIGHT_DECAY to 0.001 to prevent argmax degeneration to Neutral 2026-07-19 17:05:58 -04:00
WarriorCPU.h feat: add SGD+momentum optimizer and input-driven hyperparameters 2026-07-18 14:56:41 -04:00
WarriorCPU.lib feat: add SGD+momentum optimizer and input-driven hyperparameters 2026-07-18 14:56:41 -04:00
WarriorCPU.obj fix: lower WEIGHT_DECAY to 0.001 to prevent argmax degeneration to Neutral 2026-07-19 17:05:58 -04:00
WarriorCPU.pdb fix(Expert/ExpertSignalAIBase): add feature cache and fix oversampling imbalance 2026-07-14 22:49:14 -04:00
WarriorDML.cpp fix: lower WEIGHT_DECAY to 0.001 to prevent argmax degeneration to Neutral 2026-07-19 17:05:58 -04:00
WarriorDML.dll fix: lower WEIGHT_DECAY to 0.001 to prevent argmax degeneration to Neutral 2026-07-19 17:05:58 -04:00
WarriorDML.exp fix: lower WEIGHT_DECAY to 0.001 to prevent argmax degeneration to Neutral 2026-07-19 17:05:58 -04:00
WarriorDML.h feat: add SGD+momentum optimizer and input-driven hyperparameters 2026-07-18 14:56:41 -04:00
WarriorDML.lib feat: add SGD+momentum optimizer and input-driven hyperparameters 2026-07-18 14:56:41 -04:00
WarriorDML.obj fix: lower WEIGHT_DECAY to 0.001 to prevent argmax degeneration to Neutral 2026-07-19 17:05:58 -04:00