genpark-knowledge-distillation-teacher-student-tracker-skill
mcp
Uyari
Health Uyari
- No license — Repository has no license file
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Low visibility — Only 8 GitHub stars
Code Gecti
- Code scan — Scanned 2 files during light audit, no dangerous patterns found
Permissions Gecti
- Permissions — No dangerous permissions requested
Bu listing icin henuz AI raporu yok.
GenPark AI Agent Skill - Knowledge distillation teacher-student loss engine with temperature-scaled Kullback-Leibler divergence for continual model compression and capability preservation.
README.md
GenPark Knowledge Distillation Teacher Student Tracker Skill
Teacher-student knowledge distillation loss engine with temperature-scaled soft targets and KL divergence.
Explore more frameworks at GenPark and the GenPark MCP Catalog.
graph TD
A[Teacher Model] -->|Logits / T| B[Soft Targets Q]
C[Student Model] -->|Logits / T| D[Soft Predictions P]
B & D --> E[Distillation Loss = T^2 * KL(Q || P)]
C -->|Logits / 1.0| F[Hard Cross-Entropy Loss]
E & F --> G[Total L = (1-alpha)*L_hard + alpha*L_distill]
Features
- Temperature scaling parameterization.
- Analytical KL divergence computation.
- Pure Python standard library.
Yorumlar (0)
Yorum birakmak icin giris yap.
Yorum birakSonuc bulunamadi