Skip to content

CLAUDE.md

Latest commit

 

History

History
107 lines (83 loc) · 3.17 KB

CLAUDE.md

File metadata and controls

107 lines (83 loc) · 3.17 KB

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Project Overview

zerohertzLib is a comprehensive Python library providing utilities for scientific computing, AI/ML, data visualization, monitoring, and API integrations. The library supports modular installation with optional dependencies through extras like [api], [mlops], [quant], and [all].

Development Commands

Environment Setup

python -m venv venv
source venv/bin/activate
pip install .'[all]' --no-cache-dir

Code Quality

# Install style requirements
pip install -r requirements/requirements-style.txt

# Format code
ruff format .

# Lint and check code
ruff check .

# Check formatting (CI mode)
ruff format --check .

Testing

# Install test dependencies
pip install pytest pytest-cov

# Run all tests with coverage
PYTHONPATH=. pytest --cov=zerohertzLib --cov-report=xml --junitxml=junit.xml -o junit_family=legacy

# Run specific test module
PYTHONPATH=. pytest test/test_api.py

# Run specific test function
PYTHONPATH=. pytest test/test_api.py::test_discord_bot_message

Documentation

# Build Sphinx documentation
cd sphinx
python -m pip install -r requirements/requirements-docs.txt
# Documentation building scripts are in sphinx/ directory

Installation for Testing

# Required system dependency for vision module
sudo apt install python3-opencv -y

# Install library with all dependencies
pip install .'[all]'

Architecture Overview

Module Structure

  • api/: External service integrations (Discord, Slack, OpenAI, GitHub, etc.)
  • algorithm/: Data structures and algorithms
  • logging/: Enhanced logging utilities with rich formatting
  • mlops/: ML operations tools including Triton inference server support
  • monitoring/: System monitoring and resource tracking
  • plot/: Data visualization with matplotlib/seaborn
  • quant/: Quantitative finance tools and data analysis
  • util/: General utilities and helper functions
  • vision/: Computer vision utilities with OpenCV

Optional Dependencies

The library supports Python 3.10+ and uses conditional imports for modules with heavy dependencies:

  • API integrations require [api] extra
  • MLOps features require [mlops] extra
  • Quantitative finance requires [quant] extra
  • Vision module requires OpenCV system package

Command Line Tools

Several CLI tools are installed as scripts:

  • trictl: Triton inference server management
  • vert: Vision-related operations
  • grid: Grid operations for vision
  • img2gif/vid2gif: Media conversion utilities

Testing Environment

Tests require environment variables for API credentials:

  • DISCORD_WEBHOOK_URL, DISCORD_BOT_TOKEN, DISCORD_CHANNEL_ID
  • SLACK_WEBHOOK_URL, SLACK_BOT_TOKEN
  • OPENAI_API_KEY, GH_TOKEN

Test data is located in test/data/ directory.

Branch Strategy

  • master: Main production branch
  • dev-*: Development branches for feature work
  • CI/CD runs on pushes to dev branches and PRs to master

Code Style

  • Uses Ruff for formatting and linting (replaces Black, flake8, and pylint)
  • Consistent docstring format with type hints