{
  "architectures": [
    "SegformerForSemanticSegmentation"
  ],
  "classifier_dropout_prob": 0.1,
  "decoder_hidden_size": 256,
  "depths": [2, 2, 2, 2],
  "drop_path_rate": 0.1,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.0,
  "hidden_sizes": [32, 64, 160, 256],
  "id2label": {
    "0": "background",
    "1": "dense_forest",
    "2": "sparse_vegetation",
    "3": "deforested_area",
    "4": "water_body",
    "5": "urban_infrastructure"
  },
  "image_size": 512,
  "initializer_range": 0.02,
  "label2id": {
    "background": 0,
    "dense_forest": 1,
    "deforested_area": 3,
    "sparse_vegetation": 2,
    "urban_infrastructure": 5,
    "water_body": 4
  },
  "layer_norm_eps": 1e-06,
  "mlp_ratios": [4, 4, 4, 4],
  "model_type": "segformer",
  "num_attention_heads": [1, 2, 5, 8],
  "num_encoder_blocks": 4,
  "patch_sizes": [7, 3, 3, 3],
  "reshape_last_stage": true,
  "sr_ratios": [8, 4, 2, 1],
  "strides": [4, 2, 2, 2],
  "torch_dtype": "float32",
  "transformers_version": "4.30.2"
}