使用 SDK for SAP 的 Amazon Bedrock 執行期範例 ABAP - AWS SDK 程式碼範例
Anthropic ClaudeStable Diffusion

文件 AWS SDK AWS 範例 SDK 儲存庫中有更多可用的 GitHub 範例。

本文為英文版的機器翻譯版本，如內容有任何歧義或不一致之處，概以英文版為準。

使用 SDK for SAP 的 Amazon Bedrock 執行期範例 ABAP

下列程式碼範例示範如何使用 AWS SDK for SAP ABAP搭配 Amazon Bedrock 執行期來執行動作和實作常見案例。

每個範例都包含完整原始程式碼的連結，您可以在其中找到如何在內容中設定和執行程式碼的指示。

Anthropic Claude

下列程式碼範例示範如何使用叫用模型 API 將文字訊息傳送至 Anthropic Claude。

SDK for SAP ABAP
注意

還有更多 on GitHub。尋找完整範例，並了解如何在 AWS 程式碼範例儲存庫中設定和執行。

叫用 Anthropic Claude 2 基礎模型來產生文字。此範例使用 of /US2/CL_JSON 功能，這些功能可能無法在某些 NetWeaver 版本上使用。

    "Claude V2 Input Parameters should be in a format like this:
*   {
*     "prompt":"\n\nHuman:\\nTell me a joke\n\nAssistant:\n",
*     "max_tokens_to_sample":2048,
*     "temperature":0.5,
*     "top_k":250,
*     "top_p":1.0,
*     "stop_sequences":[]
*   }

    DATA: BEGIN OF ls_input,
            prompt               TYPE string,
            max_tokens_to_sample TYPE /aws1/rt_shape_integer,
            temperature          TYPE /aws1/rt_shape_float,
            top_k                TYPE /aws1/rt_shape_integer,
            top_p                TYPE /aws1/rt_shape_float,
            stop_sequences       TYPE /aws1/rt_stringtab,
          END OF ls_input.

    "Leave ls_input-stop_sequences empty.
    ls_input-prompt = |\n\nHuman:\\n{ iv_prompt }\n\nAssistant:\n|.
    ls_input-max_tokens_to_sample = 2048.
    ls_input-temperature = '0.5'.
    ls_input-top_k = 250.
    ls_input-top_p = 1.

    "Serialize into JSON with /ui2/cl_json -- this assumes SAP_UI is installed.
    DATA(lv_json) = /ui2/cl_json=>serialize(
      data = ls_input
                pretty_name   = /ui2/cl_json=>pretty_mode-low_case ).

    TRY.
        DATA(lo_response) = lo_bdr->invokemodel(
          iv_body = /aws1/cl_rt_util=>string_to_xstring( lv_json )
          iv_modelid = 'anthropic.claude-v2'
          iv_accept = 'application/json'
          iv_contenttype = 'application/json' ).

        "Claude V2 Response format will be:
*       {
*         "completion": "Knock Knock...",
*         "stop_reason": "stop_sequence"
*       }
        DATA: BEGIN OF ls_response,
                completion  TYPE string,
                stop_reason TYPE string,
              END OF ls_response.

        /ui2/cl_json=>deserialize(
          EXPORTING jsonx = lo_response->get_body( )
                    pretty_name = /ui2/cl_json=>pretty_mode-camel_case
          CHANGING  data  = ls_response ).

        DATA(lv_answer) = ls_response-completion.
      CATCH /aws1/cx_bdraccessdeniedex INTO DATA(lo_ex).
        WRITE / lo_ex->get_text( ).
        WRITE / |Don't forget to enable model access at https://console.aws.amazon.com/bedrock/home?#/modelaccess|.

    ENDTRY.

叫用 Anthropic Claude 2 基礎模型，使用 L2 高階用戶端產生文字。

    TRY.
        DATA(lo_bdr_l2_claude) = /aws1/cl_bdr_l2_factory=>create_claude_2( lo_bdr ).
        " iv_prompt can contain a prompt like 'tell me a joke about Java programmers'.
        DATA(lv_answer) = lo_bdr_l2_claude->prompt_for_text( iv_prompt ).
      CATCH /aws1/cx_bdraccessdeniedex INTO DATA(lo_ex).
        WRITE / lo_ex->get_text( ).
        WRITE / |Don't forget to enable model access at https://console.aws.amazon.com/bedrock/home?#/modelaccess|.

    ENDTRY.

叫用 Anthropic Claude 3 基礎模型，使用 L2 高階用戶端產生文字。

    TRY.
        " Choose a model ID from Anthropic that supports the Messages API - currently this is
        " Claude v2, Claude v3 and v3.5.  For the list of model ID, see:
        " https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids.html

        " for the list of models that support the Messages API see:
        " https://docs.aws.amazon.com/bedrock/latest/userguide/model-parameters-anthropic-claude-messages.html
        DATA(lo_bdr_l2_claude) = /aws1/cl_bdr_l2_factory=>create_anthropic_msg_api(
          io_bdr = lo_bdr
          iv_model_id = 'anthropic.claude-3-sonnet-20240229-v1:0' ).  " choosing Claude v3 Sonnet
        " iv_prompt can contain a prompt like 'tell me a joke about Java programmers'.
        DATA(lv_answer) = lo_bdr_l2_claude->prompt_for_text( iv_prompt = iv_prompt iv_max_tokens = 100 ).
      CATCH /aws1/cx_bdraccessdeniedex INTO DATA(lo_ex).
        WRITE / lo_ex->get_text( ).
        WRITE / |Don't forget to enable model access at https://console.aws.amazon.com/bedrock/home?#/modelaccess|.

    ENDTRY.

  • 如需 API 詳細資訊，請參閱 InvokeModel for Word Word 參考中的 Word。 AWS SDK SAP ABAP API

Stable Diffusion

下列程式碼範例示範如何叫用 Amazon Bedrock 上的 Stability.ai 穩定擴散 XL 來產生映像。

SDK for SAP ABAP
注意

還有更多 on GitHub。尋找完整範例，並了解如何在 AWS 程式碼範例儲存庫中設定和執行。

使用穩定擴散建立映像。

    "Stable Diffusion Input Parameters should be in a format like this:
*   {
*     "text_prompts": [
*       {"text":"Draw a dolphin with a mustache"},
*       {"text":"Make it photorealistic"}
*     ],
*     "cfg_scale":10,
*     "seed":0,
*     "steps":50
*   }
    TYPES: BEGIN OF prompt_ts,
             text TYPE /aws1/rt_shape_string,
           END OF prompt_ts.

    DATA: BEGIN OF ls_input,
            text_prompts TYPE STANDARD TABLE OF prompt_ts,
            cfg_scale    TYPE /aws1/rt_shape_integer,
            seed         TYPE /aws1/rt_shape_integer,
            steps        TYPE /aws1/rt_shape_integer,
          END OF ls_input.

    APPEND VALUE prompt_ts( text = iv_prompt ) TO ls_input-text_prompts.
    ls_input-cfg_scale = 10.
    ls_input-seed = 0. "or better, choose a random integer.
    ls_input-steps = 50.

    DATA(lv_json) = /ui2/cl_json=>serialize(
      data = ls_input
                pretty_name   = /ui2/cl_json=>pretty_mode-low_case ).

    TRY.
        DATA(lo_response) = lo_bdr->invokemodel(
          iv_body = /aws1/cl_rt_util=>string_to_xstring( lv_json )
          iv_modelid = 'stability.stable-diffusion-xl-v1'
          iv_accept = 'application/json'
          iv_contenttype = 'application/json' ).

        "Stable Diffusion Result Format:
*       {
*         "result": "success",
*         "artifacts": [
*           {
*             "seed": 0,
*             "base64": "iVBORw0KGgoAAAANSUhEUgAAAgAAA....
*             "finishReason": "SUCCESS"
*           }
*         ]
*       }
        TYPES: BEGIN OF artifact_ts,
                 seed         TYPE /aws1/rt_shape_integer,
                 base64       TYPE /aws1/rt_shape_string,
                 finishreason TYPE /aws1/rt_shape_string,
               END OF artifact_ts.

        DATA: BEGIN OF ls_response,
                result    TYPE /aws1/rt_shape_string,
                artifacts TYPE STANDARD TABLE OF artifact_ts,
              END OF ls_response.

        /ui2/cl_json=>deserialize(
          EXPORTING jsonx = lo_response->get_body( )
                    pretty_name = /ui2/cl_json=>pretty_mode-camel_case
          CHANGING  data  = ls_response ).
        IF ls_response-artifacts IS NOT INITIAL.
          DATA(lv_image) = cl_http_utility=>if_http_utility~decode_x_base64( ls_response-artifacts[ 1 ]-base64 ).
        ENDIF.
      CATCH /aws1/cx_bdraccessdeniedex INTO DATA(lo_ex).
        WRITE / lo_ex->get_text( ).
        WRITE / |Don't forget to enable model access at https://console.aws.amazon.com/bedrock/home?#/modelaccess|.

    ENDTRY.

叫用 Stability.ai Stable Diffusion XL 基礎模型，以使用 L2 高階用戶端產生映像。

    TRY.
        DATA(lo_bdr_l2_sd) = /aws1/cl_bdr_l2_factory=>create_stable_diffusion_xl_1( lo_bdr ).
        " iv_prompt contains a prompt like 'Show me a picture of a unicorn reading an enterprise financial report'.
        DATA(lv_image) = lo_bdr_l2_sd->text_to_image( iv_prompt ).
      CATCH /aws1/cx_bdraccessdeniedex INTO DATA(lo_ex).
        WRITE / lo_ex->get_text( ).
        WRITE / |Don't forget to enable model access at https://console.aws.amazon.com/bedrock/home?#/modelaccess|.

    ENDTRY.

  • 如需 API 詳細資訊，請參閱 InvokeModel AWS SDK for SAP ABAP 參考中的 API。