完成嵌入請求和回應結構描述

完成同步結構描述

{
    "schemaVersion": "nova-multimodal-embed-v1",
    "taskType": "SINGLE_EMBEDDING",
    "singleEmbeddingParams": {
        "embeddingPurpose": "GENERIC_INDEX" | "GENERIC_RETRIEVAL" | "TEXT_RETRIEVAL" | "IMAGE_RETRIEVAL" | "VIDEO_RETRIEVAL" | "DOCUMENT_RETRIEVAL" | "AUDIO_RETRIEVAL" | "CLASSIFICATION" | "CLUSTERING",
        "embeddingDimension": 256 | 384 | 1024 | 3072,
        "text": {
            "truncationMode": "START" | "END" | "NONE",
            "value": string,
            "source": SourceObject,
        },
        "image": {
            "detailLevel": "STANDARD_IMAGE" | "DOCUMENT_IMAGE",
            "format": "png" | "jpeg" | "gif" | "webp",
            "source": SourceObject
        },
        "audio": {
            "format": "mp3" | "wav" | "ogg",
            "source": SourceObject
        },
        "video": {
            "format": "mp4" | "mov" | "mkv" | "webm" | "flv" | "mpeg" | "mpg" | "wmv" | "3gp",
            "source": SourceObject,
            "embeddingMode": "AUDIO_VIDEO_COMBINED" | "AUDIO_VIDEO_SEPARATE"
        }
    }
}

   

下列清單包含請求的所有參數：

InvokeModel 回應主體

當 InvokeModel 傳回成功結果時，回應的主體將具有下列結構：

{
   "embeddings": [
      {
          "embeddingType": "TEXT" | "IMAGE" | "VIDEO" | "AUDIO" | "AUDIO_VIDEO_COMBINED",
          "embedding": number[],
          "truncatedCharLength": int // Only included if text input was truncated
      }
    ]                       
}
   

下列清單包含回應的所有參數：

完成非同步結構描述

您可以使用 Amazon Bedrock 執行時期 API 函數 StartAsyncInvoke、GetAsyncInvoke 和 ListAsyncInvokes 以非同步方式產生嵌入。如果您想要使用 Nova Embeddings 來為長內容分段，例如較長文字段落或超過 30 秒的影片和音訊，則必須使用非同步 API。

呼叫 StartAsyncInvoke 時，必須提供 modelId、outputDataConfig 和 modelInput 參數。

response = bedrock_runtime.start_async_invoke(
    modelId="amazon.nova-2-multimodal-embeddings-v1:0",
    outputDataConfig=Data Config,
    modelInput=Model Input
)
  

outputDataConfig 指定您要儲存所產生輸出的 S3 儲存貯體。其結構如下：

{
    "s3OutputDataConfig": {
        "s3Uri": "s3://your-s3-bucket"
    }
} 
  

s3Uri 是目的地儲存貯體的 S3 URI。如需其他選用參數，請參閱 StartAsyncInvoke 説明文件。

下列結構用於 modelInput 參數。

{
    "schemaVersion": "nova-multimodal-embed-v1",
    "taskType": "SEGMENTED_EMBEDDING",
    "segmentedEmbeddingParams": {
        "embeddingPurpose": "GENERIC_INDEX" | "GENERIC_RETRIEVAL" | "TEXT_RETRIEVAL" | "IMAGE_RETRIEVAL" | "VIDEO_RETRIEVAL" | "DOCUMENT_RETRIEVAL" | "AUDIO_RETRIEVAL" | "CLASSIFICATION" | "CLUSTERING",
        "embeddingDimension": 256 | 384 | 1024 | 3072,
        "text": {
            "truncationMode": "START" | "END" | "NONE",
            "value": string,
            "source": {
                "s3Location": {
                    "uri": "s3://Your S3 Object"
                }
            },
            "segmentationConfig": {
                "maxLengthChars": int
            }
        },
        "image": {
            "format": "png" | "jpeg" | "gif" | "webp",
            "source": SourceObject,
            "detailLevel": "STANDARD_IMAGE" | "DOCUMENT_IMAGE"
        },
        "audio": {
            "format": "mp3" | "wav" | "ogg",
            "source": SourceObject,
            "segmentationConfig": {
                "durationSeconds": int
            }
        },
        "video": {
            "format": "mp4" | "mov" | "mkv" | "webm" | "flv" | "mpeg" | "mpg" | "wmv" | "3gp",
            "source": SourceObject,
            "embeddingMode": "AUDIO_VIDEO_COMBINED" | "AUDIO_VIDEO_SEPARATE",
            "segmentationConfig": {
                "durationSeconds": int
            }
        }
    }
}
  

下列清單包含請求的所有參數：

StartAsyncInvoke 回應

呼叫 StartAsyncInvoke 的回應將具有以下結構。invocationArn 可以使用 GetAsyncInvoke 函數來查詢非同步任務的狀態。

{
    "invocationArn": "arn:aws:bedrock:us-east-1:xxxxxxxxxxxx:async-invoke/lvmxrnjf5mo3",
}
  

非同步輸出

當非同步嵌入產生完成時，輸出成品會寫入您指定為輸出目的地的 S3 儲存貯體。這些檔案將會有以下結構：

   amzn-s3-demo-bucket/
    job-id/
        segmented-embedding-result.json
        embedding-audio.jsonl
        embedding-image.json
        embedding-text.jsonl
        embedding-video.jsonl
        manifest.json
  

segmented-embedding-result.json 將包含整體任務結果以及對應 jsonl 檔案的參考，該檔案中包含每個模態的實際嵌入。以下是檔案的截斷範例：

{
    "sourceFileUri": string, 
    "embeddingDimension": 256 | 384 | 1024 | 3072,
    "embeddingResults": [
        {
            "embeddingType": "TEXT" | "IMAGE" | "VIDEO" | "AUDIO" | "AUDIO_VIDEO_COMBINED",
            "status": "SUCCESS" | "FAILURE" | "PARTIAL_SUCCESS",
            "failureReason": string, // Granular error codes
            "message": string, // Human-readbale failure message
            "outputFileUri": string // S3 URI to a "embedding-modality.jsonl" file
        }
        ...
    ]
}
  

embedding-modality.json 將為 jsonl 檔案，其中包含每個模態的嵌入輸出。jsonl 檔案中的每一行都會遵循下列結構描述：

{
    "embedding": number[], // The generated embedding vector
    "segmentMetadata": {
        "segmentIndex": number,
        "segmentStartCharPosition": number, // Included for text only
        "segmentEndCharPosition": number, // Included for text only
        "truncatedCharLength": number, // Included only when text gets truncated
        "segmentStartSeconds": number, // Included for audio/video only
        "segmentEndSeconds": number // Included for audio/video only
    },
    "status": "SUCCESS" | "FAILURE",
    "failureReason": string, // Granular error codes
    "message": string // Human-readable failure message
}
  

下列清單包含回應的所有參數。對於文字字元或音訊/影片時間，所有起始和結尾時間都以零為基礎。此外，所有結尾文字位置或音訊/影片時間值均包含在內。

Nova Embeddings 的檔案限制

同步作業可以同時接受 S3 輸入和內嵌區塊。非同步作業只能接受 S3 輸入。

以非同步方式產生嵌入時，您需要確保您的檔案分成適當數量的區段。對於文字嵌入，區段數量不能超過 1900 個。對於音訊和影片嵌入，區段數量不能超過 1434 個。