paramorph 0.10.0

Inephany client library to use Paramorph Agents.

Paramorph Client Library

Paramorph is a client library that provides automated hyperparameter tuning for neural network training. It integrates seamlessly with Hugging Face Transformers and other PyTorch-based training frameworks to dynamically adjust learning rates, weight decay, and other hyperparameters during training.

Features

• Automated Hyperparameter Tuning: Dynamically adjusts learning rates, weight decay, and other optimizer parameters
• Hugging Face Integration: Built-in support for Hugging Face Transformers with minimal code changes
• Multi-Agent Architecture: Uses specialized agents for different parameter groups (embeddings, attention, linear layers, convolutions)
• Real-time Monitoring: Integrates with Weights & Biases for experiment tracking
• Flexible Configuration: Easy-to-use YAML configuration system

Installation

Prerequisites

• Python 3.10+
• PyTorch
• Hugging Face Transformers (for HF integration)
• [Optional] Weights & Biases account and API key (for experiment tracking and logging)

Setup

Paramorph depends on the libinephany package, which provides core utilities and data models. Installation instructions differ based on your use case:

Ensure that python3.12 and make is installed:

Ubuntu / Debian

sudo add-apt-repository ppa:deadsnakes/ppa
sudo apt update
sudo apt install python3.12

MacOS with brew

brew install python@3.12

For Developers (Monorepo)

If you're working within the Inephany monorepo, the libinephany package is already available and will be installed into the venv created for this package when you run make install-dev.

For Clients (Standalone Installation)

Both libinephany and paramorph are available on PyPI and can be installed directly:

Optional but recommended

python -m pip install --upgrade pip

Install paramorph (libinephany will be installed as a depencency)

python -m pip install paramorph

For development installations with additional dependencies:

pip install libinephany[dev] paramorph[dev]

Then generate an API key in the portal and export it:

export PARAMORPH_API_KEY=YOUR_API_KEY

Getting Help

• Check the example scripts in the repository
• Review the configuration file format
• Ensure all dependencies are installed correctly
• Verify your model architecture matches the agent module mapping

License

This package is licensed under the Apache License, Version 2.0. See the LICENSE file for details.