dinoct

DINO-style self-supervised pretraining for OCT B-scan images, plus curve head (LoRA) post-train stage.

Repo layout

• dinoct/: Python package (models, data, training)
• configs/: YAML configs (merged: configs/ssl_default_config.yaml + configs/train/oct.yaml)
• scripts/: entrypoints and utilities

Quick Start

• Python: >=3.12
• CUDA: single GPU

This repo uses uv:

uv sync
uv run python -m dinoct --help

Data

Default expected layout under data/oct/:

• data/oct/raw/*.jpg
• data/oct/background/*.jpg
• data/oct/labeled/<image_stem>.txt (optional; marks an image as labeled)
• data/oct/extra/entries.npy (metadata cache; regenerated each run)

Each label file should contain either:

• 500 floats (one per column), or
• a 500×2 table (x, y) (the second column is used).

You can change the dataset paths via the dataset string: OCT:root=<root>[:extra=<extra>] (see configs/train/oct.yaml). In that string, root/extra refer to dataset directories (not the repo root). The dataset name token is case-insensitive.

Labeling (curve editor)

The interactive curve label editor requires matplotlib:

uv sync --extra label
uv run python scripts/data/curve_labeler.py --dir data/oct

Pretrain (SSL)

uv run python -m dinoct \
  --config configs/train/oct.yaml \
  --output-dir outputs/run1 \
  --steps 10000 \
  --post-train-steps 0

Outputs:

• outputs/run1/pretrain/dinov3_pretrain.pth
• outputs/run1/pretrain/train.log, metrics.csv, config_used.yaml

Post-train (curve head)

uv run python -m dinoct \
  --config configs/train/oct.yaml \
  --output-dir outputs/run1 \
  --steps 10000 \
  --post-train-steps 1000

To run post-train only (with an existing pretrain checkpoint):

uv run python -m dinoct \
  --config configs/train/oct.yaml \
  --output-dir outputs/run1 \
  --post-train-only \
  --pretrained-backbone outputs/run1/pretrain/dinov3_pretrain.pth \
  --post-train-steps 1000

Outputs:

• outputs/run1/post_train/fused_curve.pth
• outputs/run1/post_train/fused_curve_best.pth
• outputs/run1/post_train/val_summary.json (validation metrics for final + best checkpoint)

Visualizations

uv run python scripts/visualize.py \
  --mode curve \
  --curve-ckpt outputs/run1/post_train/fused_curve_best.pth \
  --input path/to/image_or_dir \
  --outdir outputs/viz

Export (TorchScript/ONNX)

uv run python scripts/export_model.py --model outputs/run1/post_train/fused_curve_best.pth --outdir exports

Dataset

Dataset are available from https://huggingface.co/datasets/rjbaw/oct and using the huggingface dataset loader.
Set train.dataset_path to OCT:hub=rjbaw/oct (or include root=... for local-first).

License

Apache-2.0.

This project includes alot of code derived from Meta Platforms, Inc. and affiliates' DINOv2 and DINOv3 repositories, licensed under the Apache License, Version 2.0.