Introduzione all'addestramento personalizzato: configurazione del progetto e dell'ambiente
Mantieni tutto organizzato con le raccolte
Salva e classifica i contenuti in base alle tue preferenze.
Questa pagina descrive la configurazione del progetto Google Cloud per utilizzare
Vertex AI e il download di codice TensorFlow per l'addestramento. Scaricherai
anche il codice per un'app web che riceve le previsioni.
Ogni pagina presuppone che tu abbia già eseguito le istruzioni delle pagine precedenti del tutorial.
Prima di iniziare
Durante questo tutorial, utilizza la Google Cloud console e
Cloud Shell per interagire con Google Cloud. In alternativa, anziché Cloud Shell, puoi utilizzare un'altra shell Bash con Google Cloud CLI installato.
Sign in to your Google Cloud account. If you're new to
Google Cloud,
create an account to evaluate how our products perform in
real-world scenarios. New customers also get $300 in free credits to
run, test, and deploy workloads.
In the Google Cloud console, on the project selector page,
select or create a Google Cloud project.
At the bottom of the Google Cloud console, a
Cloud Shell
session starts and displays a command-line prompt. Cloud Shell is a shell environment
with the Google Cloud CLI
already installed and with values already set for
your current project. It can take a few seconds for the session to initialize.
Se Cloud Shell non mostra
(PROJECT_ID)$
nel prompt (dove PROJECT_ID viene sostituito dal tuo
Google Cloud ID progetto), esegui questo comando per
configurare Cloud Shell in modo che utilizzi il tuo progetto:
gcloudconfigsetprojectPROJECT_ID
Crea un bucket Cloud Storage
Crea un bucket Cloud Storage regionale nella regione us-central1
da utilizzare per il resto di questo tutorial. Mentre segui il tutorial, utilizza il
bucket per diversi scopi:
Memorizza il codice di addestramento che Vertex AI deve utilizzare in un job di addestramento personalizzato.
Archivia gli artefatti del modello generati dal job di addestramento personalizzato.
Ospita l'app web che riceve le previsioni dall'endpoint Vertex AI.
Per creare il bucket Cloud Storage, esegui questo comando nella sessione di Cloud Shell:
Per visualizzare facoltativamente i file di codice campione, esegui questo comando:
ls-lpRhello-custom-sample
La directory hello-custom-sample contiene quattro elementi:
trainer/: una directory del codice TensorFlow Keras per l'addestramento del modello di classificazione dei fiori.
setup.py: un file di configurazione per il packaging della directory trainer/ in
una distribuzione dell'origine Python che Vertex AI può utilizzare.
function/: una directory di codice Python per una
funzione Cloud Run che può ricevere ed elaborare
le richieste di previsione da un browser web, inviarle a Vertex AI,
elaborare le risposte di previsione e inviarle di nuovo al browser.
webapp/: una directory con codice e markup per un'app web che riceve previsioni di classificazione dei fiori da Vertex AI.
[[["Facile da capire","easyToUnderstand","thumb-up"],["Il problema è stato risolto","solvedMyProblem","thumb-up"],["Altra","otherUp","thumb-up"]],[["Difficile da capire","hardToUnderstand","thumb-down"],["Informazioni o codice di esempio errati","incorrectInformationOrSampleCode","thumb-down"],["Mancano le informazioni o gli esempi di cui ho bisogno","missingTheInformationSamplesINeed","thumb-down"],["Problema di traduzione","translationIssue","thumb-down"],["Altra","otherDown","thumb-down"]],["Ultimo aggiornamento 2025-09-04 UTC."],[],[],null,["# Hello custom training: Set up your project and environment\n\nThis page walks through setting up your Google Cloud project to use\nVertex AI and downloading some TensorFlow code for training. You will\nalso download code for a web app that gets predictions.\nThis tutorial has several pages:\n\n\u003cbr /\u003e\n\n1. Setting up your project and environment.\n\n2. [Training a custom image classification\n model.](/vertex-ai/docs/tutorials/image-classification-custom/training)\n\n3. [Serving predictions from a custom image classification\n model.](/vertex-ai/docs/tutorials/image-classification-custom/serving)\n\n4. [Cleaning up your project.](/vertex-ai/docs/tutorials/image-classification-custom/cleanup)\n\nEach page assumes that you have already performed the instructions from the\nprevious pages of the tutorial.\n\nBefore you begin\n----------------\n\nThroughout this tutorial, use Google Cloud console and\n[Cloud Shell](/shell/docs) to interact with Google Cloud. Alternatively,\ninstead of Cloud Shell, you\ncan use another Bash shell with the [Google Cloud CLI](/sdk/docs) installed.\n\n- Sign in to your Google Cloud account. If you're new to Google Cloud, [create an account](https://console.cloud.google.com/freetrial) to evaluate how our products perform in real-world scenarios. New customers also get $300 in free credits to run, test, and deploy workloads.\n- In the Google Cloud console, on the project selector page,\n select or create a Google Cloud project.\n\n | **Note**: If you don't plan to keep the resources that you create in this procedure, create a project instead of selecting an existing project. After you finish these steps, you can delete the project, removing all resources associated with the project.\n\n [Go to project selector](https://console.cloud.google.com/projectselector2/home/dashboard)\n-\n [Verify that billing is enabled for your Google Cloud project](/billing/docs/how-to/verify-billing-enabled#confirm_billing_is_enabled_on_a_project).\n\n-\n\n\n Enable the Vertex AI and Cloud Run functions APIs.\n\n\n [Enable the APIs](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,cloudfunctions)\n\n- In the Google Cloud console, on the project selector page,\n select or create a Google Cloud project.\n\n | **Note**: If you don't plan to keep the resources that you create in this procedure, create a project instead of selecting an existing project. After you finish these steps, you can delete the project, removing all resources associated with the project.\n\n [Go to project selector](https://console.cloud.google.com/projectselector2/home/dashboard)\n-\n [Verify that billing is enabled for your Google Cloud project](/billing/docs/how-to/verify-billing-enabled#confirm_billing_is_enabled_on_a_project).\n\n-\n\n\n Enable the Vertex AI and Cloud Run functions APIs.\n\n\n [Enable the APIs](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,cloudfunctions)\n\n1. In the Google Cloud console, activate Cloud Shell.\n\n [Activate Cloud Shell](https://console.cloud.google.com/?cloudshell=true)\n\n\n At the bottom of the Google Cloud console, a\n [Cloud Shell](/shell/docs/how-cloud-shell-works)\n session starts and displays a command-line prompt. Cloud Shell is a shell environment\n with the Google Cloud CLI\n already installed and with values already set for\n your current project. It can take a few seconds for the session to initialize.\n2. If Cloud Shell does not display\n `(`\u003cvar translate=\"no\"\u003ePROJECT_ID\u003c/var\u003e`)$`\n in its prompt (where \u003cvar translate=\"no\"\u003ePROJECT_ID\u003c/var\u003e is replaced by your\n Google Cloud project ID), then run the following command to\n configure Cloud Shell to use your project:\n\n gcloud config set project \u003cvar translate=\"no\"\u003ePROJECT_ID\u003c/var\u003e\n\nCreate a Cloud Storage bucket\n-----------------------------\n\nCreate a regional [Cloud Storage](/storage/docs) bucket in the `us-central1`\nregion to use for the rest of this tutorial. As you follow the tutorial, use the\nbucket for several purposes:\n\n- Store training code for Vertex AI to use in a custom training job.\n- Store the model artifacts that your custom training job outputs.\n- Host the web app that gets predictions from your Vertex AI endpoint.\n\nTo create the Cloud Storage bucket, run the following command in your\nCloud Shell session: \n\n gcloud storage buckets create gs://\u003cvar translate=\"no\"\u003eBUCKET_NAME\u003c/var\u003e --project=\u003cvar translate=\"no\"\u003ePROJECT_ID\u003c/var\u003e --location=us-central1\n\nReplace the following:\n\n- \u003cvar translate=\"no\"\u003ePROJECT_ID\u003c/var\u003e: The ID of your Google Cloud project.\n- \u003cvar translate=\"no\"\u003eBUCKET_NAME\u003c/var\u003e: A name that you choose for your bucket. For example, `hello_custom_`\u003cvar translate=\"no\"\u003ePROJECT_ID\u003c/var\u003e. Learn about [requirements for bucket\n names](/storage/docs/buckets#naming).\n\nDownload sample code\n--------------------\n\nDownload sample code to use for the rest of the tutorial. \n\n gcloud storage cp gs://cloud-samples-data/ai-platform/hello-custom/hello-custom-sample-v1.tar.gz - | tar -xzv\n\nTo optionally view the sample code files, run the following command: \n\n ls -lpR hello-custom-sample\n\nThe `hello-custom-sample` directory has four items:\n\n- `trainer/`: A directory of TensorFlow Keras code for training the flower\n classification model.\n\n- `setup.py`: A configuration file for packaging the `trainer/` directory into\n a Python source distribution that Vertex AI can use.\n\n- `function/`: A directory of Python code for a\n [Cloud Run function](/functions/docs) that can receive and preprocess\n prediction requests from a web browser, send them to Vertex AI,\n process the prediction responses, and send them back to the browser.\n\n- `webapp/`: A directory with code and markup for a web app that gets flower\n classification predictions from Vertex AI.\n\nWhat's next\n-----------\n\nFollow the [next page of this tutorial](/vertex-ai/docs/tutorials/image-classification-custom/training) to run a custom\ntraining job on Vertex AI."]]
