Spaces:
Running
Running
| <html lang="en"> | |
| <head> | |
| <meta charset="UTF-8"> | |
| <meta name="viewport" content="width=device-width, initial-scale=1.0"> | |
| <title>Image Classification with Vertex AI – Step-by-Step Guide</title> | |
| <script src="https://cdn.tailwindcss.com"></script> | |
| <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.7.0/styles/atom-one-dark.min.css"> | |
| <script src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.7.0/highlight.min.js"></script> | |
| <script src="https://kit.fontawesome.com/3a5a3f1b9a.js" crossorigin="anonymous"></script> | |
| <style> | |
| .dark-mode { | |
| background-color: #1a202c; | |
| color: #f7fafc; | |
| } | |
| .dark-mode .card { | |
| background-color: #2d3748; | |
| border-color: #4a5568; | |
| } | |
| .dark-mode .navbar { | |
| background-color: #2d3748; | |
| border-color: #4a5568; | |
| } | |
| .dark-mode .footer { | |
| background-color: #2d3748; | |
| border-color: #4a5568; | |
| } | |
| .dark-mode .code-block { | |
| background-color: #282c34; | |
| } | |
| .dark-mode .section-icon { | |
| color: #63b3ed; | |
| } | |
| </style> | |
| </head> | |
| <body class="bg-gray-50 text-gray-800 font-sans"> | |
| <!-- Navigation --> | |
| <nav class="navbar bg-white shadow-sm sticky top-0 z-50"> | |
| <div class="container mx-auto px-4 py-3 flex justify-between items-center"> | |
| <div class="flex items-center space-x-2"> | |
| <i class="fas fa-robot text-blue-500 text-2xl"></i> | |
| <span class="text-xl font-bold">Vertex AI Guide</span> | |
| </div> | |
| <div class="flex items-center space-x-4"> | |
| <a href="#home" class="hover:text-blue-500">Home</a> | |
| <a href="#prerequisites" class="hover:text-blue-500">Prerequisites</a> | |
| <a href="#tutorial" class="hover:text-blue-500">Tutorial</a> | |
| <a href="#resources" class="hover:text-blue-500">Resources</a> | |
| <button id="darkModeToggle" class="p-2 rounded-full hover:bg-gray-200 dark-mode:hover:bg-gray-700"> | |
| <i class="fas fa-moon"></i> | |
| </button> | |
| </div> | |
| </div> | |
| </nav> | |
| <!-- Hero Section --> | |
| <section id="home" class="py-16 bg-gradient-to-r from-blue-50 to-indigo-50 dark-mode:from-gray-800 dark-mode:to-gray-900"> | |
| <div class="container mx-auto px-4"> | |
| <div class="max-w-4xl mx-auto text-center"> | |
| <h1 class="text-4xl md:text-5xl font-bold mb-6">Image Classification with Vertex AI</h1> | |
| <p class="text-xl mb-8">A step-by-step guide to training and deploying image classification models using Google Vertex AI AutoML Vision</p> | |
| <div class="flex justify-center space-x-4"> | |
| <a href="#tutorial" class="bg-blue-500 hover:bg-blue-600 text-white px-6 py-3 rounded-lg font-medium">Start Tutorial</a> | |
| <a href="#prerequisites" class="bg-gray-200 hover:bg-gray-300 dark-mode:bg-gray-700 dark-mode:hover:bg-gray-600 text-gray-800 dark-mode:text-gray-200 px-6 py-3 rounded-lg font-medium">Prerequisites</a> | |
| </div> | |
| </div> | |
| </div> | |
| </section> | |
| <!-- Introduction --> | |
| <section class="py-12"> | |
| <div class="container mx-auto px-4"> | |
| <div class="max-w-3xl mx-auto"> | |
| <div class="card bg-white p-8 rounded-lg shadow-sm border border-gray-200 mb-8"> | |
| <h2 class="text-2xl font-bold mb-4">Welcome to the Guide!</h2> | |
| <p class="mb-4">This tutorial is designed for developers, data scientists, and students who want to learn how to build image classification models without deep machine learning expertise.</p> | |
| <p class="mb-4">We'll use Google Vertex AI's AutoML Vision, which automates much of the model training process while still delivering high-quality results. No need to write complex neural network architectures!</p> | |
| <p>By the end of this guide, you'll be able to:</p> | |
| <ul class="list-disc pl-6 mt-2 space-y-1"> | |
| <li>Prepare image datasets for classification</li> | |
| <li>Train custom models with AutoML Vision</li> | |
| <li>Evaluate model performance</li> | |
| <li>Deploy models to production endpoints</li> | |
| <li>Make predictions using the Python SDK</li> | |
| </ul> | |
| </div> | |
| </div> | |
| </div> | |
| </section> | |
| <!-- Prerequisites --> | |
| <section id="prerequisites" class="py-12 bg-gray-50 dark-mode:bg-gray-900"> | |
| <div class="container mx-auto px-4"> | |
| <div class="max-w-4xl mx-auto"> | |
| <div class="flex items-center mb-8"> | |
| <i class="fas fa-clipboard-check section-icon text-3xl mr-4"></i> | |
| <h2 class="text-3xl font-bold">Prerequisites</h2> | |
| </div> | |
| <div class="grid md:grid-cols-2 gap-6"> | |
| <div class="card bg-white p-6 rounded-lg shadow-sm border border-gray-200"> | |
| <h3 class="text-xl font-semibold mb-3 flex items-center"> | |
| <i class="fas fa-cloud mr-2 text-blue-500"></i> Google Cloud Account | |
| </h3> | |
| <p>You'll need a Google Cloud account with billing enabled. Vertex AI is a paid service, but new users get $300 in free credits.</p> | |
| </div> | |
| <div class="card bg-white p-6 rounded-lg shadow-sm border border-gray-200"> | |
| <h3 class="text-xl font-semibold mb-3 flex items-center"> | |
| <i class="fas fa-project-diagram mr-2 text-blue-500"></i> Google Cloud Project | |
| </h3> | |
| <p>Create a new project or select an existing one in the Google Cloud Console where you'll enable the Vertex AI API.</p> | |
| </div> | |
| <div class="card bg-white p-6 rounded-lg shadow-sm border border-gray-200"> | |
| <h3 class="text-xl font-semibold mb-3 flex items-center"> | |
| <i class="fas fa-plug mr-2 text-blue-500"></i> Vertex AI API Enabled | |
| </h3> | |
| <p>Enable the Vertex AI API for your project. This can be done in the "APIs & Services" section of the Cloud Console.</p> | |
| </div> | |
| <div class="card bg-white p-6 rounded-lg shadow-sm border border-gray-200"> | |
| <h3 class="text-xl font-semibold mb-3 flex items-center"> | |
| <i class="fas fa-database mr-2 text-blue-500"></i> Cloud Storage Bucket | |
| </h3> | |
| <p>Create a Cloud Storage bucket to store your training data. The bucket should be in the same region where you'll train your model.</p> | |
| </div> | |
| <div class="card bg-white p-6 rounded-lg shadow-sm border border-gray-200"> | |
| <h3 class="text-xl font-semibold mb-3 flex items-center"> | |
| <i class="fas fa-code mr-2 text-blue-500"></i> Python Environment | |
| </h3> | |
| <p>Set up a Python environment (3.7+) with the Google Cloud SDK installed. We recommend using a virtual environment.</p> | |
| </div> | |
| <div class="card bg-white p-6 rounded-lg shadow-sm border border-gray-200"> | |
| <h3 class="text-xl font-semibold mb-3 flex items-center"> | |
| <i class="fas fa-key mr-2 text-blue-500"></i> Authentication | |
| </h3> | |
| <p>Set up authentication by creating a service account and downloading the JSON key file. Set the GOOGLE_APPLICATION_CREDENTIALS environment variable.</p> | |
| </div> | |
| </div> | |
| <div class="mt-8 card bg-white p-6 rounded-lg shadow-sm border border-gray-200"> | |
| <h3 class="text-xl font-semibold mb-3">Install Required Packages</h3> | |
| <p class="mb-4">Install the Google Cloud Vertex AI SDK and other required packages:</p> | |
| <pre class="code-block bg-gray-100 p-4 rounded-lg overflow-x-auto"><code class="language-bash">pip install google-cloud-aiplatform pandas</code></pre> | |
| </div> | |
| </div> | |
| </div> | |
| </section> | |
| <!-- Tutorial Steps --> | |
| <section id="tutorial" class="py-12"> | |
| <div class="container mx-auto px-4"> | |
| <div class="max-w-4xl mx-auto"> | |
| <div class="flex items-center mb-8"> | |
| <i class="fas fa-graduation-cap section-icon text-3xl mr-4"></i> | |
| <h2 class="text-3xl font-bold">Step-by-Step Tutorial</h2> | |
| </div> | |
| <!-- Step 1 --> | |
| <div class="card bg-white p-6 rounded-lg shadow-sm border border-gray-200 mb-8"> | |
| <div class="flex items-center mb-4"> | |
| <div class="bg-blue-500 text-white rounded-full w-8 h-8 flex items-center justify-center mr-3">1</div> | |
| <h3 class="text-2xl font-semibold">Dataset Preparation</h3> | |
| </div> | |
| <p class="mb-4">For image classification with AutoML Vision, your dataset needs to be structured in a specific way:</p> | |
| <div class="mb-4"> | |
| <h4 class="font-semibold mb-2">Folder Structure:</h4> | |
| <pre class="code-block bg-gray-100 p-4 rounded-lg overflow-x-auto"><code class="language-plaintext">gs://your-bucket-name/ | |
| ├── train/ | |
| │ ├── class1/ | |
| │ │ ├── image1.jpg | |
| │ │ ├── image2.jpg | |
| │ │ └── ... | |
| │ ├── class2/ | |
| │ │ ├── image1.jpg | |
| │ │ ├── image2.jpg | |
| │ │ └── ... | |
| │ └── ... | |
| └── test/ | |
| ├── class1/ | |
| ├── class2/ | |
| └── ...</code></pre> | |
| </div> | |
| <div class="mb-4"> | |
| <h4 class="font-semibold mb-2">Requirements:</h4> | |
| <ul class="list-disc pl-6 space-y-1"> | |
| <li>Minimum 10 images per class (100+ recommended for better performance)</li> | |
| <li>Images should be in JPEG or PNG format</li> | |
| <li>Each image should be at least 800x600 pixels</li> | |
| <li>Balance your dataset across classes</li> | |
| </ul> | |
| </div> | |
| <div> | |
| <h4 class="font-semibold mb-2">Upload to Cloud Storage:</h4> | |
| <p>Use the Google Cloud Console or gsutil command-line tool to upload your dataset:</p> | |
| <pre class="code-block bg-gray-100 p-4 rounded-lg overflow-x-auto mt-2"><code class="language-bash">gsutil -m cp -r /path/to/local/dataset gs://your-bucket-name</code></pre> | |
| </div> | |
| </div> | |
| <!-- Step 2 --> | |
| <div class="card bg-white p-6 rounded-lg shadow-sm border border-gray-200 mb-8"> | |
| <div class="flex items-center mb-4"> | |
| <div class="bg-blue-500 text-white rounded-full w-8 h-8 flex items-center justify-center mr-3">2</div> | |
| <h3 class="text-2xl font-semibold">Create a Vertex AI Dataset</h3> | |
| </div> | |
| <p class="mb-4">Now we'll create a dataset resource in Vertex AI that points to your Cloud Storage data.</p> | |
| <div class="mb-4"> | |
| <h4 class="font-semibold mb-2">Using the Python SDK:</h4> | |
| <pre class="code-block bg-gray-100 p-4 rounded-lg overflow-x-auto"><code class="language-python">from google.cloud import aiplatform | |
| # Initialize the Vertex AI client | |
| aiplatform.init(project="your-project-id", location="us-central1") | |
| # Create an image dataset | |
| dataset = aiplatform.ImageDataset.create( | |
| display_name="flowers-classification", | |
| gcs_source="gs://your-bucket-name/train/**", | |
| import_schema_uri=aiplatform.schema.dataset.ioformat.image.classification.single_label, | |
| ) | |
| print(f"Created dataset: {dataset.resource_name}")</code></pre> | |
| </div> | |
| <div> | |
| <h4 class="font-semibold mb-2">Alternative: Using the Console</h4> | |
| <ol class="list-decimal pl-6 space-y-1"> | |
| <li>Go to the Vertex AI section in Google Cloud Console</li> | |
| <li>Navigate to "Datasets" and click "Create"</li> | |
| <li>Select "Image classification (Single-label)"</li> | |
| <li>Enter a name and select your region</li> | |
| <li>Choose "Select import files from Cloud Storage" and enter your path (gs://your-bucket-name/train/**)</li> | |
| <li>Click "Create"</li> | |
| </ol> | |
| </div> | |
| </div> | |
| <!-- Step 3 --> | |
| <div class="card bg-white p-6 rounded-lg shadow-sm border border-gray-200 mb-8"> | |
| <div class="flex items-center mb-4"> | |
| <div class="bg-blue-500 text-white rounded-full w-8 h-8 flex items-center justify-center mr-3">3</div> | |
| <h3 class="text-2xl font-semibold">Train the AutoML Model</h3> | |
| </div> | |
| <p class="mb-4">With your dataset ready, you can now train an AutoML Vision model. This process will automatically:</p> | |
| <ul class="list-disc pl-6 mb-4 space-y-1"> | |
| <li>Split your data into training/validation sets</li> | |
| <li>Select the best model architecture</li> | |
| <li>Tune hyperparameters</li> | |
| <li>Train and evaluate the model</li> | |
| </ul> | |
| <div class="mb-4"> | |
| <h4 class="font-semibold mb-2">Using the Python SDK:</h4> | |
| <pre class="code-block bg-gray-100 p-4 rounded-lg overflow-x-auto"><code class="language-python"># Define training job | |
| training_job = aiplatform.AutoMLImageTrainingJob( | |
| display_name="train-flowers-classification", | |
| prediction_type="classification", | |
| multi_label=False, | |
| model_type="CLOUD", | |
| ) | |
| # Run the training job | |
| model = training_job.run( | |
| dataset=dataset, | |
| training_fraction_split=0.8, | |
| validation_fraction_split=0.1, | |
| test_fraction_split=0.1, | |
| budget_milli_node_hours=8000, # 8 compute hours | |
| disable_early_stopping=False, | |
| ) | |
| print(f"Training completed. Model: {model.resource_name}")</code></pre> | |
| </div> | |
| <div> | |
| <h4 class="font-semibold mb-2">Training Considerations:</h4> | |
| <ul class="list-disc pl-6 space-y-1"> | |
| <li><strong>Budget:</strong> More compute hours generally lead to better models (default is 8 hours)</li> | |
| <li><strong>Model Type:</strong> "CLOUD" for best accuracy, "MOBILE" for edge deployment</li> | |
| <li><strong>Monitoring:</strong> Track progress in the Vertex AI Console</li> | |
| </ul> | |
| </div> | |
| </div> | |
| <!-- Step 4 --> | |
| <div class="card bg-white p-6 rounded-lg shadow-sm border border-gray-200 mb-8"> | |
| <div class="flex items-center mb-4"> | |
| <div class="bg-blue-500 text-white rounded-full w-8 h-8 flex items-center justify-center mr-3">4</div> | |
| <h3 class="text-2xl font-semibold">Evaluate the Model</h3> | |
| </div> | |
| <p class="mb-4">After training completes, you'll want to evaluate the model's performance before deployment.</p> | |
| <div class="mb-4"> | |
| <h4 class="font-semibold mb-2">View Evaluation Metrics:</h4> | |
| <pre class="code-block bg-gray-100 p-4 rounded-lg overflow-x-auto"><code class="language-python"># Get evaluation metrics | |
| evaluation = model.evaluate() | |
| print("Model evaluation metrics:") | |
| print(f"Precision: {evaluation.metrics['precision']}") | |
| print(f"Recall: {evaluation.metrics['recall']}") | |
| print(f"F1 Score: {evaluation.metrics['f1Score']}") | |
| print(f"Confusion Matrix: {evaluation.metrics['confusionMatrix']}")</code></pre> | |
| </div> | |
| <div class="mb-4"> | |
| <h4 class="font-semibold mb-2">Key Metrics to Check:</h4> | |
| <ul class="list-disc pl-6 space-y-1"> | |
| <li><strong>Precision:</strong> Percentage of correct positive predictions</li> | |
| <li><strong>Recall:</strong> Percentage of actual positives correctly identified</li> | |
| <li><strong>F1 Score:</strong> Harmonic mean of precision and recall</li> | |
| <li><strong>Confusion Matrix:</strong> Shows performance per class</li> | |
| </ul> | |
| </div> | |
| <div> | |
| <h4 class="font-semibold mb-2">Console Visualization:</h4> | |
| <p>For a more visual evaluation, check the "Evaluate" tab in the Vertex AI Console where you can see:</p> | |
| <ul class="list-disc pl-6 space-y-1"> | |
| <li>Precision-recall curves</li> | |
| <li>Confusion matrix visualization</li> | |
| <li>Example predictions with confidence scores</li> | |
| </ul> | |
| </div> | |
| </div> | |
| <!-- Step 5 --> | |
| <div class="card bg-white p-6 rounded-lg shadow-sm border border-gray-200 mb-8"> | |
| <div class="flex items-center mb-4"> | |
| <div class="bg-blue-500 text-white rounded-full w-8 h-8 flex items-center justify-center mr-3">5</div> | |
| <h3 class="text-2xl font-semibold">Deploy the Model</h3> | |
| </div> | |
| <p class="mb-4">To make predictions, you need to deploy your model to an endpoint. This creates a scalable service that can handle prediction requests.</p> | |
| <div class="mb-4"> | |
| <h4 class="font-semibold mb-2">Using the Python SDK:</h4> | |
| <pre class="code-block bg-gray-100 p-4 rounded-lg overflow-x-auto"><code class="language-python"># Create an endpoint | |
| endpoint = aiplatform.Endpoint.create( | |
| display_name="flowers-classification-endpoint", | |
| project="your-project-id", | |
| location="us-central1", | |
| ) | |
| # Deploy the model to the endpoint | |
| endpoint.deploy( | |
| model=model, | |
| deployed_model_display_name="flowers-classification-model", | |
| traffic_percentage=100, | |
| machine_type="n1-standard-4", # Choose appropriate machine type | |
| min_replica_count=1, | |
| max_replica_count=1, | |
| ) | |
| print(f"Model deployed to endpoint: {endpoint.resource_name}")</code></pre> | |
| </div> | |
| <div> | |
| <h4 class="font-semibold mb-2">Deployment Considerations:</h4> | |
| <ul class="list-disc pl-6 space-y-1"> | |
| <li><strong>Machine Type:</strong> Choose based on expected traffic (n1-standard-2 for testing, larger for production)</li> | |
| <li><strong>Scaling:</strong> Set min/max replicas for automatic scaling</li> | |
| <li><strong>Cost:</strong> You're billed while the endpoint is running</li> | |
| <li><strong>Undeploy:</strong> Remember to undeploy when not in use to avoid charges</li> | |
| </ul> | |
| </div> | |
| </div> | |
| <!-- Step 6 --> | |
| <div class="card bg-white p-6 rounded-lg shadow-sm border border-gray-200"> | |
| <div class="flex items-center mb-4"> | |
| <div class="bg-blue-500 text-white rounded-full w-8 h-8 flex items-center justify-center mr-3">6</div> | |
| <h3 class="text-2xl font-semibold">Make Predictions</h3> | |
| </div> | |
| <p class="mb-4">With your model deployed to an endpoint, you can now make predictions on new images.</p> | |
| <div class="mb-4"> | |
| <h4 class="font-semibold mb-2">Using the Python SDK:</h4> | |
| <pre class="code-block bg-gray-100 p-4 rounded-lg overflow-x-auto"><code class="language-python">import base64 | |
| # Function to encode image | |
| def encode_image(image_path): | |
| with open(image_path, "rb") as image_file: | |
| return base64.b64encode(image_file.read()).decode("utf-8") | |
| # Example prediction | |
| image_path = "path/to/your/test_image.jpg" | |
| encoded_image = encode_image(image_path) | |
| # Make prediction | |
| prediction = endpoint.predict( | |
| instances=[{"content": encoded_image}], | |
| parameters={"confidenceThreshold": 0.5}, # Minimum confidence score | |
| ) | |
| # Process results | |
| for result in prediction.predictions: | |
| print("Predicted classes:") | |
| for i, (label, score) in enumerate(zip(result["displayNames"], result["confidences"])): | |
| print(f"{i+1}. {label}: {score:.2%}")</code></pre> | |
| </div> | |
| <div class="mb-4"> | |
| <h4 class="font-semibold mb-2">Alternative: Batch Prediction</h4> | |
| <p>For predicting on many images at once, use batch prediction:</p> | |
| <pre class="code-block bg-gray-100 p-4 rounded-lg overflow-x-auto"><code class="language-python"># Create batch prediction job | |
| batch_job = model.batch_predict( | |
| job_display_name="batch-pred-flowers", | |
| gcs_source="gs://your-bucket-name/test/**", | |
| gcs_destination_prefix="gs://your-bucket-name/predictions/", | |
| instances_format="jsonl", | |
| predictions_format="jsonl", | |
| ) | |
| print(f"Batch prediction job: {batch_job.resource_name}")</code></pre> | |
| </div> | |
| <div> | |
| <h4 class="font-semibold mb-2">Prediction Options:</h4> | |
| <ul class="list-disc pl-6 space-y-1"> | |
| <li><strong>Online Prediction:</strong> Low-latency requests to the endpoint (good for real-time applications)</li> | |
| <li><strong>Batch Prediction:</strong> Process many images at once (good for offline processing)</li> | |
| <li><strong>Confidence Threshold:</strong> Filter predictions by minimum confidence score</li> | |
| </ul> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| </section> | |
| <!-- Useful Resources --> | |
| <section id="resources" class="py-12 bg-gray-50 dark-mode:bg-gray-900"> | |
| <div class="container mx-auto px-4"> | |
| <div class="max-w-4xl mx-auto"> | |
| <div class="flex items-center mb-8"> | |
| <i class="fas fa-book section-icon text-3xl mr-4"></i> | |
| <h2 class="text-3xl font-bold">Useful Resources</h2> | |
| </div> | |
| <div class="grid md:grid-cols-2 gap-6"> | |
| <div class="card bg-white p-6 rounded-lg shadow-sm border border-gray-200"> | |
| <h3 class="text-xl font-semibold mb-3 flex items-center"> | |
| <i class="fas fa-file-alt mr-2 text-blue-500"></i> Official Documentation | |
| </h3> | |
| <ul class="space-y-2"> | |
| <li><a href="https://cloud.google.com/vertex-ai" class="text-blue-500 hover:underline" target="_blank">Vertex AI Documentation</a></li> | |
| <li><a href="https://cloud.google.com/vision/automl/docs" class="text-blue-500 hover:underline" target="_blank">AutoML Vision Documentation</a></li> | |
| <li><a href="https://cloud.google.com/python/docs/reference/aiplatform/latest" class="text-blue-500 hover:underline" target="_blank">Python SDK Reference</a></li> | |
| </ul> | |
| </div> | |
| <div class="card bg-white p-6 rounded-lg shadow-sm border border-gray-200"> | |
| <h3 class="text-xl font-semibold mb-3 flex items-center"> | |
| <i class="fas fa-video mr-2 text-blue-500"></i> Tutorials & Videos | |
| </h3> | |
| <ul class="space-y-2"> | |
| <li><a href="https://www.youtube.com/watch?v=zTz8w7Z8Q8I" class="text-blue-500 hover:underline" target="_blank">Vertex AI AutoML Vision Demo</a></li> | |
| <li><a href="https://cloud.google.com/blog/topics/developers-practitioners/getting-started-vertex-ai" class="text-blue-500 hover:underline" target="_blank">Getting Started with Vertex AI</a></li> | |
| <li><a href="https://cloud.google.com/ai-platform-unified/docs/tutorials" class="text-blue-500 hover:underline" target="_blank">Official Tutorials</a></li> | |
| </ul> | |
| </div> | |
| <div class="card bg-white p-6 rounded-lg shadow-sm border border-gray-200"> | |
| <h3 class="text-xl font-semibold mb-3 flex items-center"> | |
| <i class="fas fa-dollar-sign mr-2 text-blue-500"></i> Pricing & Quotas | |
| </h3> | |
| <ul class="space-y-2"> | |
| <li><a href="https://cloud.google.com/vertex-ai/pricing" class="text-blue-500 hover:underline" target="_blank">Vertex AI Pricing</a></li> | |
| <li><a href="https://cloud.google.com/vertex-ai/docs/general/quotas" class="text-blue-500 hover:underline" target="_blank">Service Quotas</a></li> | |
| <li><a href="https://cloud.google.com/free" class="text-blue-500 hover:underline" target="_blank">Free Tier Information</a></li> | |
| </ul> | |
| </div> | |
| <div class="card bg-white p-6 rounded-lg shadow-sm border border-gray-200"> | |
| <h3 class="text-xl font-semibold mb-3 flex items-center"> | |
| <i class="fas fa-users mr-2 text-blue-500"></i> Community Resources | |
| </h3> | |
| <ul class="space-y-2"> | |
| <li><a href="https://stackoverflow.com/questions/tagged/google-cloud-vertex-ai" class="text-blue-500 hover:underline" target="_blank">Stack Overflow</a></li> | |
| <li><a href="https://github.com/GoogleCloudPlatform/vertex-ai-samples" class="text-blue-500 hover:underline" target="_blank">GitHub Samples</a></li> | |
| <li><a href="https://groups.google.com/g/google-cloud-ai" class="text-blue-500 hover:underline" target="_blank">Google Group</a></li> | |
| </ul> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| </section> | |
| <!-- Footer --> | |
| <footer class="footer bg-white py-8 border-t border-gray-200"> | |
| <div class="container mx-auto px-4"> | |
| <div class="max-w-4xl mx-auto"> | |
| <div class="flex flex-col md:flex-row justify-between items-center"> | |
| <div class="mb-4 md:mb-0"> | |
| <div class="flex items-center space-x-2"> | |
| <i class="fas fa-robot text-blue-500 text-2xl"></i> | |
| <span class="text-xl font-bold">Vertex AI Guide</span> | |
| </div> | |
| <p class="text-gray-600 mt-2">A step-by-step tutorial for image classification with Vertex AI</p> | |
| </div> | |
| <div class="flex space-x-4"> | |
| <a href="#" class="text-gray-600 hover:text-blue-500"><i class="fab fa-github text-xl"></i></a> | |
| <a href="#" class="text-gray-600 hover:text-blue-500"><i class="fab fa-twitter text-xl"></i></a> | |
| <a href="#" class="text-gray-600 hover:text-blue-500"><i class="fab fa-linkedin text-xl"></i></a> | |
| </div> | |
| </div> | |
| <div class="mt-8 text-center text-gray-500 text-sm"> | |
| <p>This is an educational resource and not officially affiliated with Google Cloud.</p> | |
| <p class="mt-2">© 2023 Vertex AI Guide. All rights reserved.</p> | |
| </div> | |
| </div> | |
| </div> | |
| </footer> | |
| <script> | |
| // Initialize syntax highlighting | |
| document.addEventListener('DOMContentLoaded', (event) => { | |
| document.querySelectorAll('pre code').forEach((block) => { | |
| hljs.highlightElement(block); | |
| }); | |
| }); | |
| // Dark mode toggle | |
| const darkModeToggle = document.getElementById('darkModeToggle'); | |
| const html = document.documentElement; | |
| // Check for saved user preference | |
| const userPrefersDark = window.matchMedia && window.matchMedia('(prefers-color-scheme: dark)').matches; | |
| const currentTheme = localStorage.getItem('theme'); | |
| if (currentTheme === 'dark' || (!currentTheme && userPrefersDark)) { | |
| html.classList.add('dark-mode'); | |
| darkModeToggle.innerHTML = '<i class="fas fa-sun"></i>'; | |
| } | |
| darkModeToggle.addEventListener('click', () => { | |
| if (html.classList.contains('dark-mode')) { | |
| html.classList.remove('dark-mode'); | |
| localStorage.setItem('theme', 'light'); | |
| darkModeToggle.innerHTML = '<i class="fas fa-moon"></i>'; | |
| } else { | |
| html.classList.add('dark-mode'); | |
| localStorage.setItem('theme', 'dark'); | |
| darkModeToggle.innerHTML = '<i class="fas fa-sun"></i>'; | |
| } | |
| }); | |
| // Smooth scrolling for anchor links | |
| document.querySelectorAll('a[href^="#"]').forEach(anchor => { | |
| anchor.addEventListener('click', function (e) { | |
| e.preventDefault(); | |
| document.querySelector(this.getAttribute('href')).scrollIntoView({ | |
| behavior: 'smooth' | |
| }); | |
| }); | |
| }); | |
| </script> | |
| <p style="border-radius: 8px; text-align: center; font-size: 12px; color: #fff; margin-top: 16px;position: fixed; left: 8px; bottom: 8px; z-index: 10; background: rgba(0, 0, 0, 0.8); padding: 4px 8px;">Made with <img src="https://enzostvs-deepsite.hf.space/logo.svg" alt="DeepSite Logo" style="width: 16px; height: 16px; vertical-align: middle;display:inline-block;margin-right:3px;filter:brightness(0) invert(1);"><a href="https://enzostvs-deepsite.hf.space" style="color: #fff;text-decoration: underline;" target="_blank" >DeepSite</a> - 🧬 <a href="https://enzostvs-deepsite.hf.space?remix=mohpython/mohpython-vertex-ai" style="color: #fff;text-decoration: underline;" target="_blank" >Remix</a></p></body> | |
| </html> |