Training and evaluation

The template separates training and evaluation into two entry points:

train.py
evaluate.py

Run training

pixi run train

Equivalent manual command:

pixi run python train.py --config configs/default.yaml

Training will:

• load configs/default.yaml,
• create the data module,
• create the model,
• train with PyTorch Lightning,
• track metrics and parameters with Aim,
• save checkpoints under local/checkpoints/,
• save figures under local/figures/.

Training outputs

Generated outputs are written under:

local/

This folder is ignored by git, so experiments, checkpoints, and figures do not pollute the repository history.

Evaluate the default checkpoint

pixi run evaluate

By default, evaluation uses:

local/checkpoints/best.ckpt

Run training first unless you already have that checkpoint.

Evaluate a different checkpoint

pixi run python evaluate.py --config configs/default.yaml --ckpt path/to/checkpoint.ckpt

Evaluation logs test metrics and tracks a predicted-vs-true fit plot in Aim.

Main files involved

File	Purpose
train.py	Training entry point
evaluate.py	Evaluation entry point
configs/default.yaml	Experiment configuration
model/dataset.py	Data module
model/model.py	Neural network model
model/pl_model.py	Lightning module
callbacks.py	Aim figure logging callback