Affinage · Guide
Guide
How a domain dataset and a pinned base model become a customised local model, with a record of every step.
À venir This page is an outline. Each section below keeps its coming soon stamp until it describes code that exists in the repository.
Available now: install from source
There is no release yet. The scaffold installs from a clone of the repository:
git clone https://github.com/fonderie-dev/fonderie-affinage.git
cd fonderie-affinage
python -m pip install -e ".[dev]"
python -m pytest
The tests use tiny synthetic models and data; they need neither a GPU nor downloads.
Planned chapters
Installation
Coming soonInstalling a released version, and the optional extras that bring in the training libraries only when they are needed.
A first run
Coming soonA complete fine-tuning run on a tiny model and synthetic data, on CPU: from configuration to exported model in a few minutes.
Run configuration
Coming soonThe configuration file that describes a run: base model and revision, data, method, hyperparameters and seeds, kept under version control.
Preparing domain data
Coming soonTurning a domain dataset into training input with a manifest and checksums, and recording the train–test split that Essai checks for overlap.
Training methods
Coming soonFull fine-tuning and parameter-efficient methods such as LoRA: what each needs in memory and time, and when to choose which.
The run record
Coming soonWhat Affinage writes next to every output, following the Fonderie run record.
Export as a local model
Coming soonMerged
safetensorsweights, an adapter, or a quantised file for a local runtime, and how to choose between them.The model card
Coming soonThe generated model card: base model, data, record, licence obligations and known limitations.
Licences and restricted data
Coming soonChecking base-model and data licences before training and before release, and why weights trained on restricted data are not published. See licensing and memorisation.