ML Experiment Tracker

СреднийdataМинимальный контекст: 64K

Sets up and manages machine learning experiment tracking, hyperparameter optimization, and model registry workflows. Integrates with MLflow, Weights & Biases, and Optuna for systematic experimentation. Handles metric logging, artifact storage, model versioning, reproducibility, and automated hyperparameter search with early stopping.

Варианты использования

  • Setting up MLflow tracking for team ML projects
  • Automated hyperparameter search with Optuna
  • Comparing model performance across experiments
  • Building reproducible training pipelines with config management
  • Model versioning and promotion workflows (staging → production)
  • Generating experiment reports with metrics and visualizations

Пример промпта

Set up a complete ML experiment tracking workflow for a text classification project.

Requirements:
1. MLflow tracking server configuration (local or remote)
2. Experiment structure with proper naming and tagging
3. Optuna hyperparameter optimization with:
   - Learning rate, batch size, model architecture search
   - Pruning of unpromising trials (MedianPruner)
   - 50 trials with TPE sampler
4. Automatic logging of:
   - Training/validation metrics per epoch
   - Confusion matrix and classification report
   - Model artifacts and config files
   - Git commit hash and environment info
5. Model registry with stage transitions
6. Comparison dashboard code

Generate complete Python code using MLflow + Optuna + PyTorch Lightning.

Рекомендуемые модели

Совместимые инструменты

claude-codecursorkiroany

Модальности

Вход: text, code
Выход: text, code

Похожие Skills

Автор

OpenModels Community

@openmodelsrun