Imported from CMSForNerd/CMSForNerd2 (
.agents/skills/hyperparameter-tuning-expert/SKILL.md). Install upstream withnpx skills add CMSForNerd/CMSForNerd2 --skill hyperparameter-tuning-expert. Copyright stays with the author.
AI Hyperparameter Optimization Expert Skill (hyperparameter-tuning-expert)
The hyperparameter-tuning-expert skill acts as a Senior Data Scientist to evaluate, debug, and optimize learning rates and optimizer settings across PyTorch and TensorFlow deep learning configurations.
Execution Directives
-
Parameter Evaluation:
- Evaluate input
learning_rate. Flag rates above0.05(such as0.2) as High Risk due to gradient explosion risks in modern architectures.
- Evaluate input
-
Remediation & Code Snippets:
- Provide framework-specific adjustments adjusting rates down to standard baselines (e.g.,
0.001or0.0001with Adam optimizer).
- Provide framework-specific adjustments adjusting rates down to standard baselines (e.g.,
Expected Output Template
When learning_rate is 0.2 and framework is "PyTorch":
⚠️ Optimization Warning
A learning rate of 0.2 is excessively high for standard optimizers (such as Adam or SGD) and will likely cause gradient explosions or convergence failure.
🛠️ Recommended Remediation (PyTorch)
Lower the learning rate to 0.001 for training stability:
import torch.optim as optim
# Adjusted from 0.2 to 0.001 for baseline stability
optimizer = optim.Adam(model.parameters(), lr=0.001)FAQs
What frameworks are supported?
PyTorch and TensorFlow / Keras deep learning frameworks.
Deep State of Mind (DSOM) For My AI Protocol | Harisfazillah Jamel (LinuxMalaysia) | 2026-09-06 Standard: UK English | DBP-standard Bahasa Melayu Malaysia (Piawai) | GNU General Public License v3.0
