Skip to content

Quick Start

Use this page for the shortest path from setup to first train/inference run.

Environment & Dependencies

  • Python: 3.12.3
  • Install runtime dependencies: 
/work/envs/depth/bin/python -m pip install -r requirements.txt

Quick Training

OSTIA + Argo disk training (dataset.core.dataset_variant="ostia_argo_disk"): 

python train.py \  
  --data-config configs/px_space/data_ostia_argo_disk.yaml \  
  --train-config configs/px_space/training_config.yaml \  
  --model-config configs/px_space/model_config.yaml

Latent diffusion workflow: 

/work/envs/depth/bin/python train_autoencoder.py \  
  --ae-config configs/lat_space/ae_config.yaml \  
  --data-config configs/lat_space/data_config.yaml \  
  --train-config configs/lat_space/training_config.yaml  

/work/envs/depth/bin/python train.py \  
  --data-config configs/lat_space/data_config.yaml \  
  --train-config configs/lat_space/training_config.yaml \  
  --model-config configs/lat_space/model_config.yaml

Equivalent script wrappers:
- ./scripts/train_autoencoder.sh
- ./scripts/train_latent_diffusion.sh

See Autoencoder + Latent Diffusion for architecture, goals, limitations, and workflow details.

Quick Inference

Set config/checkpoint constants at the top of inference.py, then run: 

python inference.py

For EO multiband runs, use:
- MODEL_CONFIG_PATH = "configs/px_space/model_config.yaml"
- DATA_CONFIG_PATH = "configs/px_space/data_ostia_argo_disk.yaml"
- TRAIN_CONFIG_PATH = "configs/px_space/training_config.yaml"
Remember to wire through your dataloaders in the config. Alternatively, pass the inputs individually through PL's predict_step.