config.json

{
  "architectures": [
    "SiglipForImageClassification"
  ],
  "id2label": {
    "0": "\"rotate anticlockwise\"",
    "1": "\"increase\"",
    "2": "\"release\"",
    "3": "\"switch\"",
    "4": "\"look up\"",
    "5": "\"Terminate\"",
    "6": "\"decrease\"",
    "7": "\"move backward\"",
    "8": "\"point\"",
    "9": "\"rotate clockwise\"",
    "10": "\"grasp\"",
    "11": "\"pause\"",
    "12": "\"move forward\"",
    "13": "\"Confirm\"",
    "14": "\"look down\"",
    "15": "\"move left\"",
    "16": "\"move right\""
  },
  "initializer_factor": 1.0,
  "label2id": {
    "\"Confirm\"": 13,
    "\"Terminate\"": 5,
    "\"decrease\"": 6,
    "\"grasp\"": 10,
    "\"increase\"": 1,
    "\"look down\"": 14,
    "\"look up\"": 4,
    "\"move backward\"": 7,
    "\"move forward\"": 12,
    "\"move left\"": 15,
    "\"move right\"": 16,
    "\"pause\"": 11,
    "\"point\"": 8,
    "\"release\"": 2,
    "\"rotate anticlockwise\"": 0,
    "\"rotate clockwise\"": 9,
    "\"switch\"": 3
  },
  "model_type": "siglip",
  "problem_type": "single_label_classification",
  "text_config": {
    "attention_dropout": 0.0,
    "hidden_act": "gelu_pytorch_tanh",
    "hidden_size": 768,
    "intermediate_size": 3072,
    "layer_norm_eps": 1e-06,
    "max_position_embeddings": 64,
    "model_type": "siglip_text_model",
    "num_attention_heads": 12,
    "num_hidden_layers": 12,
    "projection_size": 768,
    "torch_dtype": "float32",
    "vocab_size": 256000
  },
  "torch_dtype": "float32",
  "transformers_version": "4.50.2",
  "vision_config": {
    "attention_dropout": 0.0,
    "hidden_act": "gelu_pytorch_tanh",
    "hidden_size": 768,
    "image_size": 224,
    "intermediate_size": 3072,
    "layer_norm_eps": 1e-06,
    "model_type": "siglip_vision_model",
    "num_attention_heads": 12,
    "num_channels": 3,
    "num_hidden_layers": 12,
    "patch_size": 16,
    "torch_dtype": "float32"
  }
}