watchtower-sdk 0.1.0

Python SDK for Watchtower AI - Data Drift & Quality Monitoring

Watchtower AI SDK

Official Python SDK for Watchtower AI - A comprehensive data drift and quality monitoring platform.

Installation

pip install watchtower-sdk

Usage

1. Configuration

Set your environment variables (recommended):

export WATCHTOWER_API_KEY="your_project_api_key"
export WATCHTOWER_API_URL="https://your-watchtower-instance.onrender.com"

2. Monitoring Data

import pandas as pd
from watchtower.monitor import WatchtowerInputMonitor

# Initialize monitor (reads env vars automatically)
monitor = WatchtowerInputMonitor(project_name="Credit Scoring Model")

# Load your data
df = pd.read_csv("production_data.csv")

# Log data to Watchtower
response = monitor.log(df)
print(response)

Features

• Data Drift Detection: Automatically detects distribution shifts in your input features.
• Data Quality Checks: Validates schema, null values, and duplicates.
• Low Latency: Async-compatible logging.