AI Blocks › Build and ship software
Model training builder
Describe your data and goal. Get training code with data loading, checks and saved checkpoints.
Model training builder is a free AI skill for the job "Build and ship software". Give the instructions to an AI assistant that supports the required files and tools, then review its output.
Free · No account · MIT licence · Made by Burhan Sebin · Level: expert
Made by Burhan Sebin. See the original on GitHub; we copied it as it is, with its licence.
What it does
Training code for your data and goal that loads the data, checks it, trains and saves progress as it goes, so a crash does not cost you the whole run. The skill profiles your data first: class balance, missing values, odd types. It splits into training, validation and test sets, keeping class ratios when they are skewed. It builds preprocessing you can reuse at prediction time, chooses a model that fits the task, such as gradient boosting for tables or a pretrained network for images, and sets the optimizer, loss and learning rate schedule. It works with PyTorch, TensorFlow or scikit-learn and adds mixed precision on GPUs.
You give it
- Your dataset or where it lives
- What you want to predict
- Your framework and hardware
- A score target (optional)
- Spreadsheet (optional)
You get
- Data loading and splitting code
- A reusable preprocessing step
- A training loop with validation
- Saved checkpoints
- Training and test scores
Use it when
- You have a labelled dataset and no training code yet.
- Your model scores well on training data and badly on new data.
- Training takes hours and you need checkpoints.
- You want preprocessing that works the same at prediction time.
How to use it
- Add it to your AI once: download it and add the folder to Claude, ChatGPT or Codex as a skill.
- Ask for the job in your own words. Your AI opens the skill when it fits.
- It works best when your AI can run code and read your project files.