{
  "dataset": "LARGEST-GLA",
  "checkpoint_type": "pretrained",
  "framework": "PyTorch",
  "hf_repo_prefix": "emelle/STGFormer-pretrain"
}