What are the best free data science tutorial sites for learning experiment tracking and fine-tuning workflows?

Here are some of the best free tutorial sites and learning resources for experiment tracking and fine-tuning workflows in data science / ML:

Top free resources

1. Hugging Face Course

• Best for: Fine-tuning modern NLP / vision / LLM workflows
• Why it’s great: Hands-on, code-first, and widely used in industry
• Covers:
  • Transformers
  • Datasets
  • Fine-tuning with Trainer / Accelerate
  • Model evaluation and deployment basics
• Link: https://huggingface.co/learn

2. MLflow Docs + Tutorials

• Best for: Experiment tracking, model registry, reproducible ML pipelines
• Why it’s great: One of the most common tools for tracking runs, parameters, metrics, and artifacts
• Covers:
  • Logging experiments
  • Comparing runs
  • Model packaging
  • Registry and deployment
• Link: https://mlflow.org/docs/latest/

3. Weights & Biases (W&B) Learn

• Best for: Experiment tracking, dashboards, sweeps, model monitoring
• Why it’s great: Very practical tutorials and examples
• Covers:
  • Logging metrics and artifacts
  • Hyperparameter sweeps
  • Visualization
  • Reports and experiment comparison
• Link: https://wandb.ai/site/learn

4. DeepLearning.AI Short Courses

• Best for: Quick practical introductions to fine-tuning and workflow tools
• Why it’s great: Free, well-structured, and often focused on current tooling
• Topics often include:
  • Fine-tuning LLMs
  • Prompt engineering
  • RAG
  • Evaluation
• Link: https://www.deeplearning.ai/short-courses/

5. Google ML Crash Course

• Best for: Core ML workflow understanding
• Why it’s great: Great for fundamentals before moving into tracking/fine-tuning
• Covers:
  • Training and evaluation
  • Overfitting
  • Feature engineering
  • Debugging ML systems
• Link: https://developers.google.com/machine-learning/crash-course

6. Kaggle Learn

• Best for: Beginner-friendly, hands-on practice
• Why it’s great: Interactive notebooks and quick tutorials
• Covers:
  • Python, pandas, ML basics
  • Intro to deep learning
  • Feature engineering
• Link: https://www.kaggle.com/learn

7. PyTorch Tutorials

• Best for: Custom fine-tuning workflows and deep learning internals
• Why it’s great: Excellent for understanding the mechanics behind training
• Covers:
  • Training loops
  • Transfer learning
  • Distributed training basics
• Link: https://pytorch.org/tutorials/

8. TensorFlow / Keras Guides

• Best for: Fine-tuning with TensorFlow ecosystem
• Why it’s great: Strong official examples for transfer learning and training pipelines
• Covers:
  • Fine-tuning pretrained models
  • Custom training
  • Callbacks and checkpoints
• Link: https://www.tensorflow.org/tutorials

---

Best picks by goal

• If you want experiment tracking:
  Start with MLflow and Weights & Biases

• If you want fine-tuning LLMs:
  Start with Hugging Face Course

• If you want both together in practice:
  Use:

  1. Hugging Face Course
  2. W&B Learn
  3. MLflow docs

• If you want fundamentals first:
  Google ML Crash Course + Kaggle Learn

---

A good free learning path

1. Google ML Crash Course or Kaggle Learn for basics
2. Hugging Face Course for fine-tuning
3. MLflow for run tracking and reproducibility
4. W&B for experiment visualization and sweeps
5. PyTorch tutorials for deeper customization

If you want, I can also give you:

• a ranked list for beginners vs. advanced learners, or
• a 1-month free study plan for experiment tracking + fine-tuning.