csviverdeia/terramind-ndvi-tokenizer

TerraMind 1.0 NDVI Tokenizer (IBM+ESA) - FSQ-VAE encoder/decoder pra imagens NDVI

Public

3 runs

Run csviverdeia/terramind-ndvi-tokenizer with an API

Use one of our client libraries to get started quickly. Clicking on a library will take you to the Playground tab where you can tweak different inputs, see the results, and copy the corresponding code to use in your own project.

Input schema

The fields you can use to run this model with an API. If you don't give a value for a field its default value will be used.

Field	Type	Default value	Description
image	string		Imagem NDVI (grayscale PNG/JPG ou GeoTIFF). Valores [0-255] mapeados pra NDVI [-1, 1].
mode	None	reconstruct	Modo de operação: 'tokenize' (encode em tokens), 'reconstruct' (encode→decode), 'tokens_only' (só índices).
timesteps	integer	10 Min: 1 Max: 50	Diffusion steps no decoder (mais steps = melhor reconstrução, mais lento).
image_size	integer	224 Min: 64 Max: 1024	Resolução de entrada (deve ser múltiplo de patch size).

{
  "type": "object",
  "title": "Input",
  "required": [
    "image"
  ],
  "properties": {
    "mode": {
      "enum": [
        "tokenize",
        "reconstruct",
        "tokens_only"
      ],
      "type": "string",
      "title": "mode",
      "description": "Modo de opera\u00e7\u00e3o: 'tokenize' (encode em tokens), 'reconstruct' (encode\u2192decode), 'tokens_only' (s\u00f3 \u00edndices).",
      "default": "reconstruct",
      "x-order": 1
    },
    "image": {
      "type": "string",
      "title": "Image",
      "format": "uri",
      "x-order": 0,
      "description": "Imagem NDVI (grayscale PNG/JPG ou GeoTIFF). Valores [0-255] mapeados pra NDVI [-1, 1]."
    },
    "timesteps": {
      "type": "integer",
      "title": "Timesteps",
      "default": 10,
      "maximum": 50,
      "minimum": 1,
      "x-order": 2,
      "description": "Diffusion steps no decoder (mais steps = melhor reconstru\u00e7\u00e3o, mais lento)."
    },
    "image_size": {
      "type": "integer",
      "title": "Image Size",
      "default": 224,
      "maximum": 1024,
      "minimum": 64,
      "x-order": 3,
      "description": "Resolu\u00e7\u00e3o de entrada (deve ser m\u00faltiplo de patch size)."
    }
  }
}

Output schema

The shape of the response you’ll get when you run this model with an API.

Schema

{
  "type": "string",
  "title": "Output"
}