Your First Steps in Dataloop 🎯

What Can You Do? 🌟

With our powerful Python SDK, you'll have full control over your entire AI development lifecycle:

Core Resources 📊

📁 Projects - Your high-level workspaces
📊 Datasets - Your data collections
🖼️ Items - Your individual files
✏️ Annotations - Your data labels
📝 Metadata - Your custom data attributes

Advanced Features 🚀

🤖 Model Management - Train, deploy, and monitor ML models
🔄 Pipelines - Build automated AI workflows
⚡ FaaS (Function as a Service) - Deploy serverless functions
📦 DPKs (Dataloop Package Kit) - Package and share your solutions
🎯 Tasks - Manage annotation and review workflows

AI Development Tools 🧠

🔍 Model Zoo - Access pre-trained models
🎓 Transfer Learning - Fine-tune existing models
📈 Model Metrics - Track performance and metrics
🔄 Data Versioning - Manage dataset versions

Automation & Integration 🔗

🔄 Webhooks - Set up event-driven workflows
🌐 REST API - Integrate with external systems
🔌 Plugins - Extend platform functionality
🤝 Team Collaboration - Manage users and roles

Authentication & Security 🔐

import dtlpy as dl

# Smart login with token handling
if dl.token_expired():
    dl.login()

# Test your installation
print(dl.__version__)

2. Token Management

# Check token status
is_expired = dl.token_expired()

# Logout
dl.logout()

3. Headless Authentication

from dotenv import load_dotenv
import os

# Load environment variables from .env file
load_dotenv()

# Access your API key securely
api_key = os.getenv('DTLPY_API_KEY')

# Initialize Dataloop with the API key
dl.login_api_key(api_key=api_key)

Project Management Mastery 🏗️

1. Creating Your First Project

# Project Setup

# Set your project name here
project_name = "onboarding-project"
try:
    # Try to get existing project
    project = dl.projects.get(project_name=project_name)
    print(f"Project '{project_name}' already exists")
except dl.exceptions.NotFound:
    # Create project if it doesn't exist
    project = dl.projects.create(project_name=project_name)
    print(f"Created project '{project_name}'")

2. Project Configuration

# Add project members
project.add_member(
    email='teammate@company.com',
    role=dl.MemberRole.DEVELOPER
)

3. Project Organization

# List all projects
projects = dl.projects.list()
for p in projects:
    print(f"Project: {p.name}")

4. Project Exploration

open_in_web() 🌐

The open_in_web() method launches the Dataloop web platform in your default browser, providing direct access to view and manage your project resources from the SDK.

# Print project details
project.print()

# Open project in Dataloop platform and explore it
project.open_in_web()

Dataset Organization 📊

1. Creating Datasets

# Dataset Setup

# Set your dataset name
dataset_name = "onboarding-dataset"

try:
    # Try to get existing dataset
    dataset = project.datasets.get(dataset_name=dataset_name)
    print(f"Dataset '{dataset_name}' already exists")
except dl.exceptions.NotFound:
    # Create dataset if it doesn't exist
    dataset = project.datasets.create(dataset_name=dataset_name)
    print(f"Created dataset '{dataset_name}'")

2. Dataset Management

# List all datasets
datasets = project.datasets.list()

# Get dataset by name
dataset = project.datasets.get(dataset_name=dataset_name)

# Print dataset details
dataset.print()

# Explore dataset in Dataloop platform
dataset.open_in_web()

Team Collaboration Essentials 👥

1. Role Management

# Add team member with role
project.add_member(
    email='annotator@company.com',
    role=dl.MemberRole.ANNOTATOR
)

# Update member role
project.update_member(
    email='annotator@company.com',
    role=dl.MemberRole.DEVELOPER
)

2. Access Control

# List project members
members = project.list_members()
for member in members:
    print(f"{member.email}: {member.role}")

# Remove member
project.remove_member(email='annotator@company.com')

Ready to start working with data? Let's move on to data management! 🚀