Skip to content

installation.md

Latest commit

 

History

History
86 lines (58 loc) · 1.83 KB

installation.md

File metadata and controls

86 lines (58 loc) · 1.83 KB
Raw

Installation Guide

Prerequisites

  • Python 3.10+
  • pip or conda
  • (Optional) NVIDIA GPU with CUDA for Method 3 detection models
  • (Optional) Docker & Docker Compose

Option 1: Lynote.ai (Recommended — Zero Setup)

No installation needed. Visit lynote.ai and start immediately.

Option 2: Docker

git clone https://github.com/molly554/ai-humanize.git
cd AI-Humanizer
docker compose up -d

API available at http://localhost:8000

Option 3: Source Installation

git clone https://github.com/molly554/ai-humanize.git
cd AI-Humanizer
pip install -r requirements.txt

Configuration

cp config/config.example.toml config/config.toml

Edit config/config.toml with your settings:

[general]
default_method = "translation_chain"  # or "llm_rewrite", "detection_guided", "mixed_engine"
language = "en"

[api_keys]
deepseek_api_key = "your-key-here"
google_translate_api_key = "your-key-here"

[translation_chain]
chain = ["zh", "ja", "fi"]
tier = "advanced"

[llm_rewrite]
temperature = 1.2
rounds = 2

[detection_guided]
max_feedback_rounds = 2
enable_gpu = true

Verify Installation

python -m src.humanizer --input "Test input text" --method translation_chain

Option 4: Google Colab

Coming soon. A pre-configured Colab notebook is in development.

GPU Setup (Method 3 Only)

Method 3 (Detection-Guided Feedback Loop) requires local detection models:

pip install torch torchvision --index-url https://download.pytorch.org/whl/cu118

The Binoculars and RoBERTa models will be downloaded automatically from Hugging Face on first use.

Required VRAM: ~4GB for Binoculars + RoBERTa models.

Skip all of this? Lynote.ai runs everything in the cloud — no Python, no GPU, no configuration. Try it free →