https://github.com/nicolasmicaux/autohparams

Automatically create a config of hyper-parameters from global variables
https://github.com/nicolasmicaux/autohparams

hyperparameter hyperparameters logger logging

Last synced: 5 months ago
JSON representation

Automatically create a config of hyper-parameters from global variables

Host: GitHub
URL: https://github.com/nicolasmicaux/autohparams
Owner: NicolasMICAUX
Created: 2023-07-26T19:56:28.000Z (almost 2 years ago)
Default Branch: main
Last Pushed: 2024-09-24T15:37:58.000Z (7 months ago)
Last Synced: 2024-11-08T14:17:51.549Z (6 months ago)
Topics: hyperparameter, hyperparameters, logger, logging
Language: Python
Homepage: https://pypi.org/project/autohparams/
Size: 74.2 KB
Stars: 1
Watchers: 1
Forks: 0
Open Issues: 0
Metadata Files:
- Readme: README.fr.md
- Contributing: CONTRIBUTING.md
- Code of conduct: CODE_OF_CONDUCT.md

Awesome Lists containing this project

README

        

[![Contributors][contributors-shield]][contributors-url]

[![Stargazers][stars-shield]][stars-url]

[![Issues][issues-shield]][issues-url]

[![PyPi version][pypi-shield]][pypi-url]

[![Python 2][python2-shield]][python-url]

[![Python 3][python3-shield]][python-url]






  

    

  

  
AutoHparams


  


    Créez automatiquement une configuration d'hyper-paramètres à partir de variables globales!

    


    


    ·

    Signaler un bug





## À propos

Avez-vous déjà effectué une expérience de machine learning pendant des heures, juste pour vous rendre compte quelques jours plus tard que vous avez oublié de sauvegarder certains hyper-paramètres? Alors vous devez prier pour que votre mémoire soit bonne, ou tout recommencer?

AutoHparams est un outil qui sauvegarde chaque variable de la portée globale de votre code dans un dictionnaire, que vous pouvez ensuite enregistrer à l'aide de votre outil préféré. Cela permet d'éviter 80% des situations d'hyper-paramètres oubliés.

## Getting Started

AutoHparams s'utilise en une ligne.

Installez AutoHparams avec pip :

```sh

pip install autohparams

```

Importez-le dans votre code, en ajoutant cette ligne :

```python

from autohparams import get_auto_hparams

```

Pour obtenir le dictionnaire de hparamsuration, faites simplement :

```python

hparams = get_auto_hparams(globals())

```

**Advanced tip**  

Par défaut, `get_auto_hparams` ignore les variables dont le nom commence par un trait de soulignement` _`. Pratique pour filtrer les variables que vous souhaitez inclure dans la hparamsuration.

Par exemple:

```python	

lr = 0.001  # nous voulons inclure le taux d'apprentissage

bs = 64     # nous voulons inclure la taille de lot

_gpu = 0    # nous ne voulons pas inclure le GPU choisi

hparams = get_auto_hparams(globals())

```

## Usage

Vous pouvez maintenant l'utiliser dans n'importe quel cadre de votre choix, par exemple :

**Tensorboard**

```python

import tensorflow as tf

from tensorboard.plugins.hparams import api as hp

with tf.summary.create_file_writer('logs/hparam_tuning').as_default():

  hp.hparams(hparams)

```

**MLflow**

```python

import mlflow

with mlflow.start_run():

  mlflow.log_params(hparams)

```

**Weights & Biases (wandb)**

```python

import wandb

wandb.init(hparams=hparams)

```

**Comet.ml**

```python

from comet_ml import Experiment

experiment = Experiment()

experiment.log_parameters(hparams)

```

**Neptune.ai**

```python

import neptune.new as neptune

run = neptune.init()

run['parameters'] = hparams

```

**Pytorch Lightning**

```python

import pytorch_lightning as pl

trainer = pl.Trainer(logger=...)

trainer.logger.log_hyperparams(hparams)

```

**Guild AI**

```python

import guild

guild.run(hparams=hparams)

```

**Polyaxon**

```python

import polyaxon_sdk

api_client = polyaxon_sdk.ApiClient()

api_client.create_hyper_params(run_uuid='uuid-of-your-run', body=hparams)

```

**ClearML**

```python

from clearml import Task

task = Task.init()

task.set_parameters(hparams)

```

**Kubeflow**

```python

from kubeflow.metadata import metadata

store = metadata.Store()

store.log_metadata(hparams)

```

#### Encore plus concis

Si vous êtes amateur de la sorcellerie de python, vous pouvez même import autohparams et l'utiliser comme une fonction:

```python

import autohparams

config = autohparams(globals())

```

On ne peut pas faire plus simple !

(back to top)


## Contributing 

Contributions are welcome!  

### Roadmap/todo

Non-Code contribution :

| Task                     | Importance | Difficulty | Contributor on it | Description                                                                                                                                                           |

|:-------------------------|------------|------------|-------------------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------|

| [Adding documentation](https://github.com/NicolasMICAUX/autohparams/discussions/6)     | 4/5        | 1/5        | NOBODY            | Write basic tutorials with real-life scenarios, write a wiki for other contributors to better understand the functioning of the library. |

_For every todo, just click on the link to find the discussion where I describe how I would do it._  

See the [discussions](https://github.com/NicolasMICAUX/autohparams/discussions) for a full list of proposed features (and known issues).

(back to top)


### How to contribute

Contributing is an awesome way to learn, inspire, and help others. Any contributions you make are **greatly appreciated**, even if it's just about styling and best practices.

If you have a suggestion that would make this project better, please fork the repo and create a pull request.  

Don't forget to give the project a star! Thanks again!

1. Fork the Project

2. Create your Feature Branch (`git checkout -b feature/YourAmazingFeature`)

3. Commit your Changes (`git commit -m 'Add some AmazingFeature'`)

4. Push to the Branch (`git push origin feature/AmazingFeature`)

5. Open a Pull Request

## Authors

This library was created by [Nicolas MICAUX](https://github.com/NicolasMICAUX).

[contributors-shield]: https://img.shields.io/github/contributors/NicolasMICAUX/autohparams.svg?style=for-the-badge

[contributors-url]: https://github.com/NicolasMICAUX/autohparams/graphs/contributors

[stars-shield]: https://img.shields.io/github/stars/NicolasMICAUX/autohparams.svg?style=for-the-badge

[stars-url]: https://github.com/NicolasMICAUX/autohparams/stargazers

[issues-shield]: https://img.shields.io/github/issues/NicolasMICAUX/autohparams.svg?style=for-the-badge

[issues-url]: https://github.com/NicolasMICAUX/autohparams/issues

[pypi-shield]: https://img.shields.io/pypi/v/searchin.svg?style=for-the-badge

[pypi-url]: https://pypi.org/project/searchin/

[python2-shield]: https://img.shields.io/badge/python-2.7+-blue.svg?style=for-the-badge

[python3-shield]: https://img.shields.io/badge/python-3.5+-blue.svg?style=for-the-badge

[python-url]: https://www.python.org/downloads/

[//]: # ([license-shield]: https://img.shields.io/github/license/NicolasMICAUX/autohparams.svg?style=for-the-badge)

[//]: # ([license-url]: https://github.com/NicolasMICAUX/autohparams/blob/master/LICENSE.txt)

[//]: # ([linkedin-shield]: https://img.shields.io/badge/-LinkedIn-black.svg?style=for-the-badge&logo=linkedin&colorB=555)

[//]: # ([linkedin-url]: https://linkedin.com/in/othneildrew)

ecosyste.ms

Data

Tools

Indexes

Applications

Experiments

Awesome

https://github.com/nicolasmicaux/autohparams

Awesome Lists containing this project

README

AutoHparams