# model-opt

**Repository Path**: AungV/model-opt

## Basic Information

- **Project Name**: model-opt
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2026-04-14
- **Last Updated**: 2026-06-18

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# CamOpt

AI Code Optimization Tools for PyTorch/torch_npu inference services.

## Installation

```bash
# Install
uv tool install .

# Development mode
pip install -e ".[dev]"
```

## Usage

### CLI Tools

```bash
# 通路级耗时分析
camopt timeline --input ./my_inference_service

# Torch profiler 性能分析
camopt profile --input ./my_inference_service --device npu
```

### Claude Code Skills

Install skills to your Claude Code environment:

```bash
cp -r skills/* ~/.claude/skills/
```

Then use in Claude Code:

```
/camopt ./my_inference_service
```

## Development

```bash
pip install -e ".[dev]"
pytest
```

## Project Structure

```
camopt/
├── src/camopt/          # Python CLI tools
│   ├── cli.py           # CLI entry point
│   ├── profiling/       # Profiling modules
│   └── utils/           # Shared utilities
├── skills/              # Claude Code Skills
│   ├── camopt-*/        # Framework skills
│   └── opt-*/           # Optimization skills
├── tests/               # Test suite
└── examples/            # Example inference services
```