Quick Start Guide

This guide will help you quickly create your first ML workspace and run a sample project using Trailer.dev.

Prerequisites

1. Install Trailer.dev

   # Pull the standalone container
   docker pull ghcr.io/trailer-dev/trailer:latest

2. Start Trailer.dev

   # Run in standalone mode
   docker run -p 8090:8090 --gpus all ghcr.io/trailer-dev/trailer:latest

3. Access the Web Interface

   • Open your browser and navigate to http://localhost:8090
   • Log in with your credentials

Create Your First Workspace

1. Create a New Workspace

   • Click "New Workspace" in the web interface
   • Choose a name for your workspace
   • Select "Python ML" as the template

2. Configure Python Environment

3. Set Up GPU Support

4. Run a Sample Project

   # sample.py
   import torch
   from transformers import AutoModel, AutoTokenizer
   
   # Load model and tokenizer
   model = AutoModel.from_pretrained("bert-base-uncased")
   tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
   
   # Test GPU availability
   print(f"GPU available: {torch.cuda.is_available()}")
   if torch.cuda.is_available():
       print(f"GPU device: {torch.cuda.get_device_name(0)}")

5. Save and Share Workspace

Next Steps

1. Explore Features

   • Architecture Overview
   • Configuration Reference

2. Get Support

   • Documentation
   • Contact Support