[[["わかりやすい","easyToUnderstand","thumb-up"],["問題の解決に役立った","solvedMyProblem","thumb-up"],["その他","otherUp","thumb-up"]],[["わかりにくい","hardToUnderstand","thumb-down"],["情報またはサンプルコードが不正確","incorrectInformationOrSampleCode","thumb-down"],["必要な情報 / サンプルがない","missingTheInformationSamplesINeed","thumb-down"],["翻訳に関する問題","translationIssue","thumb-down"],["その他","otherDown","thumb-down"]],["最終更新日 2025-09-04 UTC。"],[],[],null,["# Hello tabular data: Deploy a model and request a prediction\n\nAfter your AutoML tabular classification model is done training,\ncreate an endpoint and deploy your model to\nthe endpoint. After your model is deployed to this new endpoint, test your\nmodel by requesting a prediction.\n\nLoad your model\n---------------\n\nWhen your model finishes training, it is listed in the **Models** tab.\n\n1. In the Google Cloud console, in the Vertex AI section, go to\n the **Models** page.\n\n [Go to the Models page](https://console.cloud.google.com/vertex-ai/models)\n2. From the models list, click the name of your trained model that you created previously\n\n3. Models are organized into versions. Click model version number 1.\n\nEvaluate your model\n-------------------\n\nThe **Evaluate** panel helps you understand how the model performed against the\ntest set. When you are done, continue to the next part of the tutorial.\n\n### Evaluation metrics\n\n**Optional** . Hold the pointer over the `?` icons to learn about each evaluation\nmetric.\n\n**Optional**. Move the confidence threshold slider to see how the precision,\nrecall, and F1 scores are affected.\n\n### Confusion matrix\n\nThe **confusion matrix** shows how a prediction compares to the test set (ground\ntruth).\n\nRecall that label \"1\" is the negative class (the customer did not sign up for a\nterm deposit) and \"2\" is the positive class. Your model likely did a better job\npredicting the negative class than the positive class. Perhaps with additional\ntraining time, more data, or additional features, you could improve predictive\nperformance for the positive class.\n\n### Feature importance\n\n**Feature importance** shows how each feature impacted model training: The\nhigher the value, the more impactful.\n\nYour model probably shows that **duration** (how long the most recent\ncommunication between the bank and customer lasted, in seconds) contributed\nheavily to the prediction outcome.\n\nDeploy your model to an endpoint\n--------------------------------\n\nTo test a model or make online predictions, you need to deploy it to an endpoint.\n\n1. Open the **Deploy \\& Test** panel.\n\n2. Under **Deploy your model** , click **Deploy to endpoint**.\n\n3. Enter `Structured_AutoML_Tutorial` for the endpoint name.\n\n4. Click **Continue**.\n\n5. Keep the minimum compute node at `1` and don't enter a maximum.\n\n6. Select `n1-standard-2` machine type.\n\n7. Click **Continue**.\n\n8. Turn off model monitoring for this endpoint.\n\n9. To create your endpoint and deploy your model to the endpoint,\n click **Deploy**.\n\n Model deployment takes around 5 minutes. When your endpoint is ready, proceed\n to the next part of the tutorial.\n | **Note:** Make sure to undeploy the model later so you aren't charged for additional compute resources. We'll show you how to do that at the end of this tutorial.\n\nRequest a prediction\n--------------------\n\nNow that your model is deployed to an endpoint, you can send prediction\nrequests. Rather than send a request through the API or gcloud, you can test\nyour model on this page.\n\n1. In the **Test your model** section, you'll see a **Value** column that's\n pre-filled. You can use those values or enter new ones.\n\n2. At the bottom of the section, press **Predict**.\n\n For this model, a prediction result of `1` represents a negative\n outcome---a deposit is not made at the bank. A prediction result of `2`\n represents a positive outcome---a deposit is made at the bank.\n\n Your model will return a confidence score, which is the model's level of\n certainty that the selected label is the correct one. The default value\n probably returned a high confidence score.\n3. **Optional** . Try changing **duration** to a much higher value and press Predict again.\n\nWhat's next\n-----------\n\n- To avoid incurring unexpected charges, follow the instructions in\n [Clean up your project](/vertex-ai/docs/tutorials/tabular-automl/cleanup)\n\n- [Learn more](/vertex-ai/docs/tabular-data/classification-regression/evaluate-model)\n about model evaluation.\n\n- [Learn more](/vertex-ai/docs/tabular-data/classification-regression/get-online-predictions)\n about model predictions."]]
