Vertex AI는 두 가지 모델 학습 방법인 AutoML과 커스텀 학습을 제공합니다. AutoML을 사용하면 최소한의 노력 및 머신러닝 환경으로 학습을 수행할 수 있고 커스텀 학습을 사용하면 학습 특성을 완전히 제어할 수 있습니다. 학습 방법을 자세히 알아보세요.
Vertex AI는 소스 데이터 유형 및 특성 값을 검사하고 모델 학습에서 해당 특성을 사용할 방법을 추론합니다. 각 열의 데이터 유형을 검토해서 올바르게 해석되었는지 확인하는 것이 좋습니다. 필요한 경우 모든 특성에 대해 지원되는 서로 다른 변환을 지정할 수 있습니다.
변환에 대해 자세히 알아보세요.
[[["이해하기 쉬움","easyToUnderstand","thumb-up"],["문제가 해결됨","solvedMyProblem","thumb-up"],["기타","otherUp","thumb-up"]],[["이해하기 어려움","hardToUnderstand","thumb-down"],["잘못된 정보 또는 샘플 코드","incorrectInformationOrSampleCode","thumb-down"],["필요한 정보/샘플이 없음","missingTheInformationSamplesINeed","thumb-down"],["번역 문제","translationIssue","thumb-down"],["기타","otherDown","thumb-down"]],["최종 업데이트: 2025-09-08(UTC)"],[],[],null,["# Hello tabular data: Create a dataset and train an AutoML classification model\n\nUse the Google Cloud console to create a tabular dataset and train a classification\nmodel.\n\nCreate a tabular dataset\n------------------------\n\n1. In the Google Cloud console, in the Vertex AI section, go to\n the **Datasets** page.\n\n [Go to the Datasets page](https://console.cloud.google.com/vertex-ai/datasets)\n2. Click **Create** in the button bar to create a new dataset.\n\n3. Enter `Structured_AutoML_Tutorial` for the dataset name and select the\n **Tabular** tab.\n\n4. Select the **Regression/Classification** objective.\n\n Leave the **Region** set to **us-central1**.\n5. Click **Create** to create the dataset.\n\n For this tutorial, you'll use a publicly available bank dataset hosted on Cloud Storage.\n6. For **Select a data source** , click **Select CSV files from Cloud\n Storage**\n\n7. In **Import file path** , enter `cloud-ml-tables-data/bank-marketing.csv`\n\n8. Click **Continue**.\n\nAnalyze the dataset\n-------------------\n\nThe analyze section lets you view more information about the dataset, like\nmissing or NULL values.\n\nBecause our dataset is formatted correctly for this tutorial, you don't need to\ndo anything on this page and can skip this section.\n\n1. **Optional** . Click **Generate statistics** to view the number of missing or\n NULL values in the dataset. This can take 10 minutes or longer.\n\n2. **Optional**. Click on one of the feature columns to learn more about the data values.\n\nTrain an AutoML classification model\n------------------------------------\n\n1. Click **Train new model**.\n\n2. Select **Other**.\n\n3. In the **Training method** pane, confirm that the dataset you created\n previously is selected for the **Dataset** field.\n\n4. For the **Objective** field, select **Classification**.\n\n5. Confirm that the AutoML training method is selected.\n\n6. Click **Continue**.\n\n7. In the **Model details** pane, select **Deposit** for the target column and\n click **Continue**.\n\n The target column is what we're training the model to predict. For the\n `bank-marketing.csv` dataset, the `Deposit` column indicates whether the\n client purchased a term deposit (2 = yes, 1 = no).\n\n The **Training options** pane gives you an opportunity to add features and\n transform column data. If no columns are selected, then by default all non-\n target columns will be used as features for training. This dataset is ready\n to use, so there's no need to apply any transformations.\n8. Click **Continue**.\n\n9. In the **Compute and pricing** pane, enter `1` for the training budget.\n\n The training budget is the maximum time (may vary slightly) that the model\n spends training. This value is multiplied by the\n [price per node hour](/vertex-ai/pricing#automl_models)\n to calculate to total training cost. More training hours results in a more\n accurate (up to a point) model but results in a higher cost. For development\n purposes, a low budget is fine but for production it's important to strike a\n balance between cost and accuracy.\n10. Click **Start training**.\n\nWhen the model finishes training, it's displayed in the model tab as a live\nlink, with a green checkmark status icon.\n\nWhat's next\n-----------\n\nYour model is now being trained, which can take an hour or more to complete.\nYou'll receive an email when training is complete. When your model has finished\ntraining, follow the [next page of this tutorial](/vertex-ai/docs/tutorials/tabular-automl/deploy-predict) to\ndeploy your model and request a prediction.\n\nThis tutorial uses a dataset that's been cleaned and formatted for\nAutoML training, but most data will require some work before it's ready\nto be used. The quality of your training data impacts the effectiveness of the\nmodels you create. [Learn more](/vertex-ai/docs/tabular-data/classification-regression/prepare-data) about preparing data.\n\nSourcing and preparing your data is a critical to ensuring an accurate machine\nlearning model. [Learn more](/vertex-ai/docs/tabular-data/bp-tabular) about best\npractices.\n\n[Learn more](/vertex-ai/docs/tabular-data/classification-regression/create-dataset)\nabout creating a tabular dataset.\n\nVertex AI offers two model training methods, AutoML and custom\ntraining. AutoML lets you train with minimal effort and machine\nlearning experience, while custom training gives you complete control over\ntraining functionality. [Learn more](/vertex-ai/docs/start/training-methods) about\ntraining methods.\n\nVertex AI examines the source data type and feature values and infers how\nit will use that feature in model training. It's recommended that you review\neach column's data type to verify that it's been interpreted correctly. If\nneeded, you can specify a different supported transformation for any feature.\n[Learn more](/vertex-ai/docs/datasets/data-types-tabular) about transformations.\n\n[Learn more](/vertex-ai/docs/tabular-data/classification-regression/train-model) about training an AutoML for classification or regression."]]