POST /transform/summarize

{
  "body": {
    "config": {
      "length": 3
    }
  },
  "parameters": {},
  "input_asset": {
    "assetId": "019007f3-b9e1-704a-addf-7996adab6409",
    "assetType": "post"
  },
  "side_effects": true
}

{
  "post": {
    "name": "Summary of \"TimeGPT model methodology question\"",
    "content": {
      "json": {
        "type": "doc",
        "content": [
          {
            "type": "heading",
            "attrs": {
              "level": 1
            },
            "content": [
              {
                "text": "Summary of \"TimeGPT model methodology question\"",
                "type": "text"
              }
            ]
          },
          {
            "type": "paragraph",
            "content": [
              {
                "text": "By using embeddings to capture the contextual meaning of time series data, TimeGPT can distinguish between similar series like seasonal versus promotional sales or residential versus industrial energy consumption. This approach allows the model to make more accurate predictions by recognizing unique patterns and characteristics specific to each context. The latent space created by embeddings also enables the model to generalize patterns across similar contexts, differentiate diverse contexts, learn dependencies and interactions, and incorporate external knowledge for a richer understanding of time series data.",
                "type": "text"
              }
            ]
          }
        ]
      },
      "text": "Summary of \"TimeGPT model methodology question\"\nBy using embeddings to capture the contextual meaning of time series data, TimeGPT can distinguish between similar series like seasonal versus promotional sales or residential versus industrial energy consumption. This approach allows the model to make more accurate predictions by recognizing unique patterns and characteristics specific to each context. The latent space created by embeddings also enables the model to generalize patterns across similar contexts, differentiate diverse contexts, learn dependencies and interactions, and incorporate external knowledge for a richer understanding of time series data."
    },
    "description": ""
  }
}