Skip to content

Commit

Permalink
Merge branch 'master' of github.com:DLHub-Argonne/dlhub_sdk
Browse files Browse the repository at this point in the history
  • Loading branch information
WardLT committed Oct 31, 2018
2 parents 3103e09 + 8e7554f commit 363b3eb
Showing 1 changed file with 4 additions and 6 deletions.
10 changes: 4 additions & 6 deletions README.md
Original file line number Diff line number Diff line change
@@ -1,19 +1,17 @@
# DLHub SDK
[![Build Status](https://travis-ci.org/DLHub-Argonne/dlhub_toolbox.svg?branch=master)](https://travis-ci.org/DLHub-Argonne/dlhub_toolbox)[![Coverage Status](https://coveralls.io/repos/github/DLHub-Argonne/dlhub_toolbox/badge.svg?branch=master)](https://coveralls.io/github/DLHub-Argonne/dlhub_toolbox?branch=master)
[![Build Status](https://travis-ci.org/DLHub-Argonne/dlhub_sdk.svg?branch=master)](https://travis-ci.org/DLHub-Argonne/dlhub_sdk)[![Coverage Status](https://coveralls.io/repos/github/DLHub-Argonne/dlhub_sdk/badge.svg?branch=master)](https://coveralls.io/github/DLHub-Argonne/dlhub_sdk?branch=master)[![PyPI version](https://badge.fury.io/py/dlhub-sdk.svg)](https://badge.fury.io/py/dlhub-sdk)

DLHub SDK contains scripts designed to make it easier to submit datasets and machine learning models to the Data and Learning Hub for Science (DLHub).
This package provides utility functions for building data in the correct schema for DLHub and tools that automatically extract metadata from common types of datasets and ML models.

## Installation

DLHub SDK is on PyPi, and can be installed using pip

```
git clone https://github.com/DLHub-Argonne/dlhub_toolbox.git
cd dlhub_toolbox
pip install -e .
pip install dlhub-sdk
```

`dlhub_sdk` is not yet on PyPi. So, you have to install it by first cloning the repository and then calling `pip install -e .`

## Example Usage

As a simple example, we will show how to submit a machine learning model created based on the [Iris Dataset](https://archive.ics.uci.edu/ml/datasets/Iris).
Expand Down

0 comments on commit 363b3eb

Please sign in to comment.