Automobile Insurance Claim Processing with Azure OpenAI > Using the Automobile Insurance Claim Processing with Azure OpenAI recipe > Step 4. Configure and publish the processes
  

Step 4. Configure and publish the processes

Configure the vehicle details, LLM model ID, API version, and generation configuration details, and then publish the processes.
    1Open the Validate Vehicle Information process.
    2On the Assignments tab of the Save Sample Vehicle Dataset step, in the VehicleDataset field, you can change the vehicles data or add an HTTP request to get external data.
    3Save and publish the process.
    4Open the Get Price List and Parts List process.
    5On the Assignments tab of the Save Sample Price Dataset step, in the PriceDataset field, you can change the vehicles parts price list or add an HTTP request to get external data.
    6Save and publish the process.
    7To publish the Check Image Format process, click Actions in the row that contains the process and select Publish.
    8Open the Damage Recognition using Azure OpenAI LLM process.
    9On the Temp Fields tab of the Start step, in the Retry field, you can change the value to increase the number of attempts to call the LLM model if error an occurs. By default, it is set to 1.
    10On the Assignments tab of the Prompt Request step, configure the following properties:
    1. aUpdate the GenerationConfig field using the Expression Editor, as shown in the following sample code:
    2. <generationConfig>
      <temperature>0.3</temperature>
      <maxOutputTokens>200</maxOutputTokens>
      <topP>0.8</topP>
      </generationConfig>
      For the GenerationConfig field, enter values for the following properties:
      temperature
      Controls the randomness of the model's output. A lower value makes the output more deterministic, while a higher value increases randomness and creativity. For example, a temperature of 0.5 balances between deterministic and creative outputs.
      maxOutputTokens
      Defines the maximum number of tokens the model can generate in its response. Setting a limit ensures that the response is concise and fits within the desired length constraints.
      topP
      Determines the cumulative probability threshold for token selection. The model considers the smallest set of tokens whose cumulative probability meets or exceeds topP. For example, if topP is set to 0.1, the model considers only the top 10% most probable tokens at each step.
    3. bEnter the model deployment ID in the Deployment_ID field.
    4. cEnter the API version in the Api_Version field. The API version must use the YYYY-MM-DD or YYYY-MM-DD-preview format.
    11Save and publish the process.
    12To publish the Calculating the Payout process, click Actions in the row that contains the process and select Publish.
    13To publish the Send Email with Claim Details process, click Actions in the row that contains the process and select Publish.
    14To publish the Process Claim Request process, click Actions in the row that contains the process and select Publish.