Installation

This guide walks you through installing and configuring the ViewAI Python SDK.

Prerequisites

Before installing the ViewAI SDK, ensure you have:

• Python 3.8 or higher installed

• pip package manager

• Virtual environment (recommended)

• ViewAI API key (get one at app.viewai.ca)

Check Python Version

python --version

# or
python3 --version

If you don't have Python 3.8+, download it from python.org.

Installation Methods

Method 1: Install from PyPI (Recommended)

The easiest way to install the ViewAI SDK is from PyPI:

pip install viewai-client

Method 2: Install from Source

For development or to use the latest features:

# Clone the repository
git clone https://github.com/viewai/viewai-client-python.git
cd viewai-client-python

# Install in development mode
pip install -e .

Method 3: Install with Optional Dependencies

For ONNX model support:

pip install viewai-client[onnx]

For development with all dependencies:

pip install viewai-client[dev]

Virtual Environment Setup

Using venv (Recommended)

# Create virtual environment
python -m venv viewai-env

# Activate on Linux/macOS
source viewai-env/bin/activate

# Activate on Windows
viewai-env\Scripts\activate

# Install ViewAI SDK
pip install viewai-client

Using conda

# Create conda environment
conda create -n viewai-env python=3.10

# Activate environment
conda activate viewai-env

# Install ViewAI SDK
pip install viewai-client

Verify Installation

After installation, verify everything is working:

import viewai_client
print(f"ViewAI SDK version: {viewai_client.__version__}")

Or run a quick test:

from viewai_client import ViewAIClient

# This should not raise any errors
client = ViewAIClient(api_key="test-key")
print("ViewAI SDK installed successfully!")

Dependencies

The SDK automatically installs the following required dependencies:

Core Dependencies

• requests (>=2.28.0) - HTTP client

• pandas (>=1.5.0) - Data manipulation

• numpy (>=1.21.0) - Numerical computing

• tqdm (>=4.64.0) - Progress bars

Optional Dependencies

• scikit-learn (>=1.1.0) - Model training

• onnxruntime - ONNX model inference

• skl2onnx - Model export to ONNX

• joblib - Model serialization

Configuration

Environment Variables

Set up your API key and configuration using environment variables:

# Linux/macOS
export VIEWAI_API_KEY="your-api-key-here"
export VIEWAI_BASE_URL="https://api.viewai.ca"
export VIEWAI_TIMEOUT="30"

# Windows (Command Prompt)
set VIEWAI_API_KEY=your-api-key-here
set VIEWAI_BASE_URL=https://api.viewai.ca
set VIEWAI_TIMEOUT=30

# Windows (PowerShell)
$env:VIEWAI_API_KEY="your-api-key-here"
$env:VIEWAI_BASE_URL="https://api.viewai.ca"
$env:VIEWAI_TIMEOUT="30"

Configuration File

Create a .env file in your project root:

# .env
VIEWAI_API_KEY=your-api-key-here
VIEWAI_BASE_URL=https://api.viewai.ca
VIEWAI_TIMEOUT=30
VIEWAI_MAX_RETRIES=3

Load it using python-dotenv:

from dotenv import load_dotenv
load_dotenv()

from viewai_client import ViewAIClient
client = ViewAIClient()  # Uses env variables

Platform-Specific Instructions

Linux

# Install pip if not available
sudo apt-get update
sudo apt-get install python3-pip

# Install ViewAI SDK
pip3 install viewai-client

macOS

Windows

Docker Installation

For containerized environments:

Dockerfile

FROM python:3.10-slim

# Set working directory
WORKDIR /app

# Install ViewAI SDK
RUN pip install --no-cache-dir viewai-client

# Copy your application
COPY . .

# Set environment variables
ENV VIEWAI_API_KEY=""

# Run your application
CMD ["python", "your_script.py"]

docker-compose.yml

version: '3.8'

services:
  viewai-app:
    build: .
    environment:
      - VIEWAI_API_KEY=${VIEWAI_API_KEY}
      - VIEWAI_BASE_URL=https://api.viewai.ca
    volumes:
      - ./data:/app/data

Jupyter Notebook Setup

For use in Jupyter notebooks:

# Install Jupyter
pip install jupyter

# Install ViewAI SDK
pip install viewai-client

# Start Jupyter
jupyter notebook

In your notebook:

# Install directly in notebook (if needed)
!pip install viewai-client

# Import and use
from viewai_client import ViewAIClient
client = ViewAIClient(api_key="your-api-key")

Google Colab Setup

For Google Colab:

# Install ViewAI SDK
!pip install viewai-client

# Import and configure
from viewai_client import ViewAIClient
from google.colab import userdata

# Get API key from Colab secrets
api_key = userdata.get('VIEWAI_API_KEY')
client = ViewAIClient(api_key=api_key)

Troubleshooting

Common Issues

ModuleNotFoundError: No module named 'viewai_client'

Solution: Ensure you've activated your virtual environment and installed the SDK:

pip install viewai-client

ImportError: cannot import name 'ViewAIClient'

Solution: Check your Python version (must be 3.8+):

python --version

SSL Certificate errors

Solution: Update certifi:

pip install --upgrade certifi

Permission denied errors on Linux/macOS

Solution: Use user installation:

pip install --user viewai-client

Dependency Conflicts

If you encounter dependency conflicts:

# Create a fresh virtual environment
python -m venv fresh-env
source fresh-env/bin/activate

# Install ViewAI SDK
pip install viewai-client

# Freeze dependencies
pip freeze > requirements.txt

Network Issues

If installation fails due to network issues:

# Use a different PyPI mirror
pip install --index-url https://pypi.org/simple/ viewai-client

# Or increase timeout
pip install --timeout=300 viewai-client

Upgrading

Upgrade to Latest Version

pip install --upgrade viewai-client

Upgrade to Specific Version

pip install viewai-client==1.2.3

Check Installed Version

import viewai_client
print(viewai_client.__version__)

Uninstallation

To remove the ViewAI SDK:

pip uninstall viewai-client

Development Setup

For contributing to the SDK:

# Clone repository
git clone https://github.com/viewai/viewai-client-python.git
cd viewai-client-python

# Create virtual environment
python -m venv dev-env
source dev-env/bin/activate

# Install in editable mode with dev dependencies
pip install -e ".[dev]"

# Install pre-commit hooks
pre-commit install

# Run tests
pytest tests/

Next Steps

1

Get Your API Key

Visit app.viewai.ca to create an account and get your API key.

2

Quick Start Tutorial

Follow the Quick Start Guide to build your first model.

3

Authentication Setup

Learn about Authentication.

4

Core Concepts

Understand Core Concepts.

Additional Resources

• PyPI Package

• GitHub Repository

• Changelog

• Support

Need Help? Contact us at [email protected] or visit our community forum.