No Description

Tolstoyevsky 9e74ba23d5 Higher LR and batch size 1 week ago
.dvc fe46adaa97 Initial commit 1 week ago
dagshub fe46adaa97 Initial commit 1 week ago
examples 9e74ba23d5 Higher LR and batch size 1 week ago
tests fe46adaa97 Initial commit 1 week ago
.gitignore fe46adaa97 Initial commit 1 week ago
.travis.yml fe46adaa97 Initial commit 1 week ago
LICENSE fe46adaa97 Initial commit 1 week ago
README.md fe46adaa97 Initial commit 1 week ago
dagshub_github.png fe46adaa97 Initial commit 1 week ago
requirements.txt fe46adaa97 Initial commit 1 week ago
setup.py fe46adaa97 Initial commit 1 week ago

Data Pipeline

Legend
DVC Managed File
Git Managed File
Metric
Stage File
External File

README.md



Build Status pypi

DAGsHub Python client libraries

Use DAGsHub to create reproducible versions of your data science research project, allow others to understand your project, and to contribute back to it.

DAGsHub is built firmly around open, standard formats for your project. In particular:

  • git
  • DVC
  • Standard data formats like YAML, JSON, CSV

Therefore, you can work with DAGsHub regardless of your chosen programming language or frameworks.

This client library is meant to help you get started quickly in Python, but it's purely optional - the data formats are very simple and you can choose to work with them directly.

You can learn more by completing our short tutorial or reading the docs

Installation

pip install dagshub

Basic Usage

from dagshub import dagshub_logger, DAGsHubLogger

# As a context manager:
with dagshub_logger() as logger:
    # Metrics:
    logger.log_metrics(loss=3.14, step_num=1)
    # OR:
    logger.log_metrics({'val_loss': 6.28}, step_num=2)
    
    # Hyperparameters:
    logger.log_hyperparams(lr=1e-4)
    # OR:
    logger.log_hyperparams({'optimizer': 'sgd'})
    

# As a normal Python object:
logger = DAGsHubLogger()
logger.log_hyperparams(num_layers=32)
logger.log_metrics(batches_per_second=100, step_num=42)
# ...
logger.save()
logger.close()

Integrations with ML frameworks

The basic DAGsHub logger is just plain Python, and requires no specific framework.

However, for convenience, we include some integrations with common ML frameworks, which can just work right out of the box, without having to write any logging code on your own:


Made with 🐶 by DAGsHub.