| Name | paramorph JSON |
| Version |
0.6.0
JSON |
| download |
| home_page | None |
| Summary | Inephany client library to use Paramorph Agents. |
| upload_time | 2025-07-28 16:18:49 |
| maintainer | None |
| docs_url | None |
| author | None |
| requires_python | >=3.10 |
| license | Apache 2.0 |
| keywords |
paramorph
agents
client
|
| VCS |
|
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
# 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
```commandline
sudo add-apt-repository ppa:deadsnakes/ppa
sudo apt update
sudo apt install python3.12
```
#### MacOS with brew
```commandline
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
```bash
python -m pip install --upgrade pip
```
# Install paramorph (libinephany will be installed as a depencency)
```bash
python -m pip install paramorph
```
For development installations with additional dependencies:
```bash
pip install libinephany[dev] paramorph[dev]
```
Then generate an API key in the portal and export it:
```commandline
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.
Raw data
{
"_id": null,
"home_page": null,
"name": "paramorph",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.10",
"maintainer_email": null,
"keywords": "paramorph, agents, client",
"author": null,
"author_email": "Inephany <info@inephany.com>",
"download_url": "https://files.pythonhosted.org/packages/13/c8/d2f038f5a20e6a638cf7a798c202aa653efc55c036961911e6f25de5c081/paramorph-0.6.0.tar.gz",
"platform": null,
"description": "# Paramorph Client Library\n\nParamorph 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.\n\n## Features\n\n- **Automated Hyperparameter Tuning**: Dynamically adjusts learning rates, weight decay, and other optimizer parameters\n- **Hugging Face Integration**: Built-in support for Hugging Face Transformers with minimal code changes\n- **Multi-Agent Architecture**: Uses specialized agents for different parameter groups (embeddings, attention, linear layers, convolutions)\n- **Real-time Monitoring**: Integrates with Weights & Biases for experiment tracking\n- **Flexible Configuration**: Easy-to-use YAML configuration system\n\n## Installation\n\n### Prerequisites\n\n- Python 3.10+\n- PyTorch\n- Hugging Face Transformers (for HF integration)\n- [Optional] Weights & Biases account and API key (for experiment tracking and logging)\n\n### Setup\n\nParamorph depends on the `libinephany` package, which provides core utilities and data models. Installation instructions differ based on your use case:\n\nEnsure that python3.12 and make is installed:\n\n#### Ubuntu / Debian\n```commandline\nsudo add-apt-repository ppa:deadsnakes/ppa\nsudo apt update\nsudo apt install python3.12\n```\n\n#### MacOS with brew\n```commandline\nbrew install python@3.12\n```\n\n#### For Developers (Monorepo)\nIf 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`.\n\n#### For Clients (Standalone Installation)\nBoth `libinephany` and `paramorph` are available on PyPI and can be installed directly:\n\n# Optional but recommended\n```bash\npython -m pip install --upgrade pip\n```\n\n# Install paramorph (libinephany will be installed as a depencency)\n```bash\npython -m pip install paramorph\n```\n\nFor development installations with additional dependencies:\n\n```bash\npip install libinephany[dev] paramorph[dev]\n```\n\nThen generate an API key in the portal and export it:\n```commandline\nexport PARAMORPH_API_KEY=YOUR_API_KEY\n```\n\n### Getting Help\n\n- Check the example scripts in the repository\n- Review the configuration file format\n- Ensure all dependencies are installed correctly\n- Verify your model architecture matches the agent module mapping\n\n## License\n\nThis package is licensed under the Apache License, Version 2.0. See the LICENSE file for details.\n",
"bugtrack_url": null,
"license": "Apache 2.0",
"summary": "Inephany client library to use Paramorph Agents.",
"version": "0.6.0",
"project_urls": null,
"split_keywords": [
"paramorph",
" agents",
" client"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "dffbcc593dfb95c4dfde159ddd34765c42be8e2a7fb3faea59ec520e3097b779",
"md5": "f0b0faea6b35f22beb150f2c6eb83034",
"sha256": "66af4abc652823da98462cce8f7198307d5b4a38ca26715dda7ac31edd819e94"
},
"downloads": -1,
"filename": "paramorph-0.6.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "f0b0faea6b35f22beb150f2c6eb83034",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.10",
"size": 23386,
"upload_time": "2025-07-28T16:18:48",
"upload_time_iso_8601": "2025-07-28T16:18:48.432961Z",
"url": "https://files.pythonhosted.org/packages/df/fb/cc593dfb95c4dfde159ddd34765c42be8e2a7fb3faea59ec520e3097b779/paramorph-0.6.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "13c8d2f038f5a20e6a638cf7a798c202aa653efc55c036961911e6f25de5c081",
"md5": "fc0a7cf1f45f5d80f6509303e6d20f60",
"sha256": "f3800dca7f45c8012ded1b823089c6adc6cf7cc399cc42c83c79045835e236a4"
},
"downloads": -1,
"filename": "paramorph-0.6.0.tar.gz",
"has_sig": false,
"md5_digest": "fc0a7cf1f45f5d80f6509303e6d20f60",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.10",
"size": 27094,
"upload_time": "2025-07-28T16:18:49",
"upload_time_iso_8601": "2025-07-28T16:18:49.402021Z",
"url": "https://files.pythonhosted.org/packages/13/c8/d2f038f5a20e6a638cf7a798c202aa653efc55c036961911e6f25de5c081/paramorph-0.6.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-07-28 16:18:49",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "paramorph"
}
