async function submitInference()

in src/js/app.js [89:142]
31 lines of code
5 McCabe index (conditional complexity)

async function submitInference() {

  try {
    //================================================
    // Validate the user inputs:
    validateUserInputs();

    //================================================
    // Remove any spaces from the endpointNameSelect that can happen from copy and paste.
    let epName = endpointNameSelect.value.replace(/ /g, '');

    // Update the text output div
    inferenceText.innerHTML = `Image Selected: ${imageFileSelect.files[0].name}<br />`;
    inferenceText.innerHTML += `Sagemaker Endpoint: ${epName}<br />`;

    // Get the inference classes the user entered into a list.
    classMap = [];
    for (let i = 0; i <= classCnt; i++) {
      classMap[i] = $(`#inference-class-${i}`).val();
    }
    inferenceText.innerHTML += `Inference Classes: ${JSON.stringify(classMap)}<br />`;

    inferenceText.innerHTML += `Sending the image for inference to the Sagemaker Endpoint......<br />`;

    // Post the request
    // TODO: If image source is very large can cause long delay, add option to scale image down before POST'ing
    let apiInit = {
      body: {
        endpoint: epName,
        region: regionSelect.value,
        imageBase64: originalImage.src
      }
      // headers: {} // OPTIONAL
    }
    let response = await API.post(apiName, inferenceApiPath, apiInit);

    if (!response.statusCode) {
      throw Error("An unknown error occured")
    } else if (response.statusCode !== 200) {
      throw Error(response.error_message);
    }

    // if no error, get the predictions and process.
    predictions = response.predictions;

    // Makes prediction vaues more readable and display to UI
    inferenceText.innerHTML += getPredictionHtmlPrettyPrint();
    applyPredictionsToImage();

  } catch (err) {
    inferenceText.innerHTML += `ERROR: An error occured:<br />${err}`;
    console.log(err, err.stack);
  }
}