<p align="center"><img width=50% src="https://raw.githubusercontent.com/alteryx/compose/main/docs/source/images/compose.png" alt="Compose" /></p>
<p align="center"><i>"Build better training examples in a fraction of the time."</i></p>
<p align="center">
<a href="https://github.com/alteryx/compose/actions?query=workflow%3ATests" target="_blank">
<img src="https://github.com/alteryx/compose/workflows/Tests/badge.svg" alt="Tests" />
</a>
<a href="https://codecov.io/gh/alteryx/compose">
<img src="https://codecov.io/gh/alteryx/compose/branch/main/graph/badge.svg?token=mDz4ueTUEO"/>
</a>
<a href="https://compose.alteryx.com/en/stable/?badge=stable" target="_blank">
<img src="https://readthedocs.com/projects/feature-labs-inc-compose/badge/?version=stable&token=5c3ace685cdb6e10eb67828a4dc74d09b20bb842980c8ee9eb4e9ed168d05b00"
alt="ReadTheDocs" />
</a>
<a href="https://badge.fury.io/py/composeml" target="_blank">
<img src="https://badge.fury.io/py/composeml.svg?maxAge=2592000" alt="PyPI Version" />
</a>
<a href="https://stackoverflow.com/questions/tagged/compose-ml" target="_blank">
<img src="https://img.shields.io/badge/questions-on_stackoverflow-blue.svg?" alt="StackOverflow" />
</a>
<a href="https://pepy.tech/project/composeml" target="_blank">
<img src="https://pepy.tech/badge/composeml/month" alt="PyPI Downloads" />
</a>
</p>
<hr>
[Compose](https://compose.alteryx.com) is a machine learning tool for automated prediction engineering. It allows you to structure prediction problems and generate labels for supervised learning. An end user defines an outcome of interest by writing a *labeling function*, then runs a search to automatically extract training examples from historical data. Its result is then provided to [Featuretools](https://docs.featuretools.com/) for automated feature engineering and subsequently to [EvalML](https://evalml.alteryx.com/) for automated machine learning. The workflow of an applied machine learning engineer then becomes:
<br><p align="center"><img width=90% src="https://raw.githubusercontent.com/alteryx/compose/main/docs/source/images/workflow.png" alt="Compose" /></p><br>
By automating the early stage of the machine learning pipeline, our end user can easily define a task and solve it. See the [documentation](https://compose.alteryx.com) for more information.
## Installation
Install with pip
```
python -m pip install composeml
```
or from the Conda-forge channel on [conda](https://anaconda.org/conda-forge/composeml):
```
conda install -c conda-forge composeml
```
### Add-ons
**Update checker** - Receive automatic notifications of new Compose releases
```
python -m pip install "composeml[update_checker]"
```
## Example
> Will a customer spend more than 300 in the next hour of transactions?
In this example, we automatically generate new training examples from a historical dataset of transactions.
```python
import composeml as cp
df = cp.demos.load_transactions()
df = df[df.columns[:7]]
df.head()
```
<table border="0" class="dataframe">
<thead>
<tr style="text-align: right;">
<th>transaction_id</th>
<th>session_id</th>
<th>transaction_time</th>
<th>product_id</th>
<th>amount</th>
<th>customer_id</th>
<th>device</th>
</tr>
</thead>
<tbody>
<tr>
<td>298</td>
<td>1</td>
<td>2014-01-01 00:00:00</td>
<td>5</td>
<td>127.64</td>
<td>2</td>
<td>desktop</td>
</tr>
<tr>
<td>10</td>
<td>1</td>
<td>2014-01-01 00:09:45</td>
<td>5</td>
<td>57.39</td>
<td>2</td>
<td>desktop</td>
</tr>
<tr>
<td>495</td>
<td>1</td>
<td>2014-01-01 00:14:05</td>
<td>5</td>
<td>69.45</td>
<td>2</td>
<td>desktop</td>
</tr>
<tr>
<td>460</td>
<td>10</td>
<td>2014-01-01 02:33:50</td>
<td>5</td>
<td>123.19</td>
<td>2</td>
<td>tablet</td>
</tr>
<tr>
<td>302</td>
<td>10</td>
<td>2014-01-01 02:37:05</td>
<td>5</td>
<td>64.47</td>
<td>2</td>
<td>tablet</td>
</tr>
</tbody>
</table>
First, we represent the prediction problem with a labeling function and a label maker.
```python
def total_spent(ds):
return ds['amount'].sum()
label_maker = cp.LabelMaker(
target_dataframe_index="customer_id",
time_index="transaction_time",
labeling_function=total_spent,
window_size="1h",
)
```
Then, we run a search to automatically generate the training examples.
```python
label_times = label_maker.search(
df.sort_values('transaction_time'),
num_examples_per_instance=2,
minimum_data='2014-01-01',
drop_empty=False,
verbose=False,
)
label_times = label_times.threshold(300)
label_times.head()
```
<table border="0" class="dataframe">
<thead>
<tr style="text-align: right;">
<th>customer_id</th>
<th>time</th>
<th>total_spent</th>
</tr>
</thead>
<tbody>
<tr>
<td>1</td>
<td>2014-01-01 00:00:00</td>
<td>True</td>
</tr>
<tr>
<td>1</td>
<td>2014-01-01 01:00:00</td>
<td>True</td>
</tr>
<tr>
<td>2</td>
<td>2014-01-01 00:00:00</td>
<td>False</td>
</tr>
<tr>
<td>2</td>
<td>2014-01-01 01:00:00</td>
<td>False</td>
</tr>
<tr>
<td>3</td>
<td>2014-01-01 00:00:00</td>
<td>False</td>
</tr>
</tbody>
</table>
We now have labels that are ready to use in [Featuretools](https://docs.featuretools.com/) to generate features.
## Support
The Innovation Labs open source community is happy to provide support to users of Compose. Project support can be found in three places depending on the type of question:
1. For usage questions, use [Stack Overflow](https://stackoverflow.com/questions/tagged/compose-ml) with the `composeml` tag.
2. For bugs, issues, or feature requests start a Github [issue](https://github.com/alteryx/compose/issues/new).
3. For discussion regarding development on the core library, use [Slack](https://join.slack.com/t/alteryx-oss/shared_invite/zt-182tyvuxv-NzIn6eiCEf8TBziuKp0bNA).
4. For everything else, the core developers can be reached by email at open_source_support@alteryx.com
## Citing Compose
Compose is built upon a newly defined part of the machine learning process — prediction engineering. If you use Compose, please consider citing this paper:
James Max Kanter, Gillespie, Owen, Kalyan Veeramachaneni. [Label, Segment,Featurize: a cross domain framework for prediction engineering.](https://dai.lids.mit.edu/wp-content/uploads/2017/10/Pred_eng1.pdf) IEEE DSAA 2016.
BibTeX entry:
```bibtex
@inproceedings{kanter2016label,
title={Label, segment, featurize: a cross domain framework for prediction engineering},
author={Kanter, James Max and Gillespie, Owen and Veeramachaneni, Kalyan},
booktitle={2016 IEEE International Conference on Data Science and Advanced Analytics (DSAA)},
pages={430--439},
year={2016},
organization={IEEE}
}
```
## Acknowledgements
The open source development has been supported in part by DARPA's Data driven discovery of models program (D3M).
## Alteryx
**Compose** is an open source project maintained by [Alteryx](https://www.alteryx.com). We developed Compose to enable flexible definition of the machine learning task. To see the other open source projects we’re working on visit [Alteryx Open Source](https://www.alteryx.com/open-source). If building impactful data science pipelines is important to you or your business, please get in touch.
<p align="center">
<a href="https://www.alteryx.com/open-source">
<img src="https://alteryx-oss-web-images.s3.amazonaws.com/OpenSource_Logo-01.png" alt="Alteryx Open Source" width="800"/>
</a>
</p>
Raw data
{
"_id": null,
"home_page": "",
"name": "composeml",
"maintainer": "",
"docs_url": null,
"requires_python": "<4,>=3.8",
"maintainer_email": "\"Alteryx, Inc.\" <open_source_support@alteryx.com>",
"keywords": "prediction engineering,data science,machine learning",
"author": "",
"author_email": "\"Alteryx, Inc.\" <open_source_support@alteryx.com>",
"download_url": "https://files.pythonhosted.org/packages/98/d7/70264fb178f79f6c7b1981cf6780dc0a2feb4ed76bdea771882b38b971f7/composeml-0.10.1.tar.gz",
"platform": null,
"description": "<p align=\"center\"><img width=50% src=\"https://raw.githubusercontent.com/alteryx/compose/main/docs/source/images/compose.png\" alt=\"Compose\" /></p>\n<p align=\"center\"><i>\"Build better training examples in a fraction of the time.\"</i></p>\n<p align=\"center\">\n <a href=\"https://github.com/alteryx/compose/actions?query=workflow%3ATests\" target=\"_blank\">\n <img src=\"https://github.com/alteryx/compose/workflows/Tests/badge.svg\" alt=\"Tests\" />\n </a>\n <a href=\"https://codecov.io/gh/alteryx/compose\">\n <img src=\"https://codecov.io/gh/alteryx/compose/branch/main/graph/badge.svg?token=mDz4ueTUEO\"/>\n </a>\n <a href=\"https://compose.alteryx.com/en/stable/?badge=stable\" target=\"_blank\">\n <img src=\"https://readthedocs.com/projects/feature-labs-inc-compose/badge/?version=stable&token=5c3ace685cdb6e10eb67828a4dc74d09b20bb842980c8ee9eb4e9ed168d05b00\"\n alt=\"ReadTheDocs\" />\n </a>\n <a href=\"https://badge.fury.io/py/composeml\" target=\"_blank\">\n <img src=\"https://badge.fury.io/py/composeml.svg?maxAge=2592000\" alt=\"PyPI Version\" />\n </a>\n <a href=\"https://stackoverflow.com/questions/tagged/compose-ml\" target=\"_blank\">\n <img src=\"https://img.shields.io/badge/questions-on_stackoverflow-blue.svg?\" alt=\"StackOverflow\" />\n </a>\n <a href=\"https://pepy.tech/project/composeml\" target=\"_blank\">\n <img src=\"https://pepy.tech/badge/composeml/month\" alt=\"PyPI Downloads\" />\n </a>\n</p>\n<hr>\n\n[Compose](https://compose.alteryx.com) is a machine learning tool for automated prediction engineering. It allows you to structure prediction problems and generate labels for supervised learning. An end user defines an outcome of interest by writing a *labeling function*, then runs a search to automatically extract training examples from historical data. Its result is then provided to [Featuretools](https://docs.featuretools.com/) for automated feature engineering and subsequently to [EvalML](https://evalml.alteryx.com/) for automated machine learning. The workflow of an applied machine learning engineer then becomes:\n\n<br><p align=\"center\"><img width=90% src=\"https://raw.githubusercontent.com/alteryx/compose/main/docs/source/images/workflow.png\" alt=\"Compose\" /></p><br>\n\nBy automating the early stage of the machine learning pipeline, our end user can easily define a task and solve it. See the [documentation](https://compose.alteryx.com) for more information.\n\n## Installation\nInstall with pip\n\n```\npython -m pip install composeml\n```\n\nor from the Conda-forge channel on [conda](https://anaconda.org/conda-forge/composeml):\n\n```\nconda install -c conda-forge composeml\n```\n\n### Add-ons\n\n**Update checker** - Receive automatic notifications of new Compose releases\n\n```\npython -m pip install \"composeml[update_checker]\"\n```\n\n## Example\n> Will a customer spend more than 300 in the next hour of transactions?\n\nIn this example, we automatically generate new training examples from a historical dataset of transactions.\n\n```python\nimport composeml as cp\ndf = cp.demos.load_transactions()\ndf = df[df.columns[:7]]\ndf.head()\n```\n\n<table border=\"0\" class=\"dataframe\">\n <thead>\n <tr style=\"text-align: right;\">\n <th>transaction_id</th>\n <th>session_id</th>\n <th>transaction_time</th>\n <th>product_id</th>\n <th>amount</th>\n <th>customer_id</th>\n <th>device</th>\n </tr>\n </thead>\n <tbody>\n <tr>\n <td>298</td>\n <td>1</td>\n <td>2014-01-01 00:00:00</td>\n <td>5</td>\n <td>127.64</td>\n <td>2</td>\n <td>desktop</td>\n </tr>\n <tr>\n <td>10</td>\n <td>1</td>\n <td>2014-01-01 00:09:45</td>\n <td>5</td>\n <td>57.39</td>\n <td>2</td>\n <td>desktop</td>\n </tr>\n <tr>\n <td>495</td>\n <td>1</td>\n <td>2014-01-01 00:14:05</td>\n <td>5</td>\n <td>69.45</td>\n <td>2</td>\n <td>desktop</td>\n </tr>\n <tr>\n <td>460</td>\n <td>10</td>\n <td>2014-01-01 02:33:50</td>\n <td>5</td>\n <td>123.19</td>\n <td>2</td>\n <td>tablet</td>\n </tr>\n <tr>\n <td>302</td>\n <td>10</td>\n <td>2014-01-01 02:37:05</td>\n <td>5</td>\n <td>64.47</td>\n <td>2</td>\n <td>tablet</td>\n </tr>\n </tbody>\n</table>\n\nFirst, we represent the prediction problem with a labeling function and a label maker.\n\n```python\ndef total_spent(ds):\n return ds['amount'].sum()\n\nlabel_maker = cp.LabelMaker(\n target_dataframe_index=\"customer_id\",\n time_index=\"transaction_time\",\n labeling_function=total_spent,\n window_size=\"1h\",\n)\n```\n\nThen, we run a search to automatically generate the training examples.\n\n```python\nlabel_times = label_maker.search(\n df.sort_values('transaction_time'),\n num_examples_per_instance=2,\n minimum_data='2014-01-01',\n drop_empty=False,\n verbose=False,\n)\n\nlabel_times = label_times.threshold(300)\nlabel_times.head()\n```\n\n<table border=\"0\" class=\"dataframe\">\n <thead>\n <tr style=\"text-align: right;\">\n <th>customer_id</th>\n <th>time</th>\n <th>total_spent</th>\n </tr>\n </thead>\n <tbody>\n <tr>\n <td>1</td>\n <td>2014-01-01 00:00:00</td>\n <td>True</td>\n </tr>\n <tr>\n <td>1</td>\n <td>2014-01-01 01:00:00</td>\n <td>True</td>\n </tr>\n <tr>\n <td>2</td>\n <td>2014-01-01 00:00:00</td>\n <td>False</td>\n </tr>\n <tr>\n <td>2</td>\n <td>2014-01-01 01:00:00</td>\n <td>False</td>\n </tr>\n <tr>\n <td>3</td>\n <td>2014-01-01 00:00:00</td>\n <td>False</td>\n </tr>\n </tbody>\n</table>\n\nWe now have labels that are ready to use in [Featuretools](https://docs.featuretools.com/) to generate features.\n\n## Support\n\nThe Innovation Labs open source community is happy to provide support to users of Compose. Project support can be found in three places depending on the type of question:\n\n1. For usage questions, use [Stack Overflow](https://stackoverflow.com/questions/tagged/compose-ml) with the `composeml` tag.\n2. For bugs, issues, or feature requests start a Github [issue](https://github.com/alteryx/compose/issues/new).\n3. For discussion regarding development on the core library, use [Slack](https://join.slack.com/t/alteryx-oss/shared_invite/zt-182tyvuxv-NzIn6eiCEf8TBziuKp0bNA).\n4. For everything else, the core developers can be reached by email at open_source_support@alteryx.com\n\n## Citing Compose\nCompose is built upon a newly defined part of the machine learning process \u2014 prediction engineering. If you use Compose, please consider citing this paper:\nJames Max Kanter, Gillespie, Owen, Kalyan Veeramachaneni. [Label, Segment,Featurize: a cross domain framework for prediction engineering.](https://dai.lids.mit.edu/wp-content/uploads/2017/10/Pred_eng1.pdf) IEEE DSAA 2016.\n\nBibTeX entry:\n\n```bibtex\n@inproceedings{kanter2016label,\n title={Label, segment, featurize: a cross domain framework for prediction engineering},\n author={Kanter, James Max and Gillespie, Owen and Veeramachaneni, Kalyan},\n booktitle={2016 IEEE International Conference on Data Science and Advanced Analytics (DSAA)},\n pages={430--439},\n year={2016},\n organization={IEEE}\n}\n```\n\n## Acknowledgements \n\nThe open source development has been supported in part by DARPA's Data driven discovery of models program (D3M). \n\n## Alteryx\n\n**Compose** is an open source project maintained by [Alteryx](https://www.alteryx.com). We developed Compose to enable flexible definition of the machine learning task. To see the other open source projects we\u2019re working on visit [Alteryx Open Source](https://www.alteryx.com/open-source). If building impactful data science pipelines is important to you or your business, please get in touch.\n\n<p align=\"center\">\n <a href=\"https://www.alteryx.com/open-source\">\n <img src=\"https://alteryx-oss-web-images.s3.amazonaws.com/OpenSource_Logo-01.png\" alt=\"Alteryx Open Source\" width=\"800\"/>\n </a>\n</p>\n",
"bugtrack_url": null,
"license": "BSD 3-clause",
"summary": "a framework for automated prediction engineering",
"version": "0.10.1",
"split_keywords": [
"prediction engineering",
"data science",
"machine learning"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "f509d830ebb1401a2e2ca445e22b91d338c7056a2e53776afed9fc422356a91e",
"md5": "9a9d7a8050364b91ba61834630de958f",
"sha256": "e6d292cbd8619d8e5649207be3954f1b04900b651e51ee85d838ab05d51b7bdb"
},
"downloads": -1,
"filename": "composeml-0.10.1-py3-none-any.whl",
"has_sig": false,
"md5_digest": "9a9d7a8050364b91ba61834630de958f",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<4,>=3.8",
"size": 39408,
"upload_time": "2023-01-07T03:28:32",
"upload_time_iso_8601": "2023-01-07T03:28:32.634270Z",
"url": "https://files.pythonhosted.org/packages/f5/09/d830ebb1401a2e2ca445e22b91d338c7056a2e53776afed9fc422356a91e/composeml-0.10.1-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "98d770264fb178f79f6c7b1981cf6780dc0a2feb4ed76bdea771882b38b971f7",
"md5": "acb86c6efe955e5a43dcfed52a5d6278",
"sha256": "d87fc181be72fadec16ea0e44ebd363b4e2e3620c87c8b2cae6e236fd00761d4"
},
"downloads": -1,
"filename": "composeml-0.10.1.tar.gz",
"has_sig": false,
"md5_digest": "acb86c6efe955e5a43dcfed52a5d6278",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<4,>=3.8",
"size": 32480,
"upload_time": "2023-01-07T03:28:34",
"upload_time_iso_8601": "2023-01-07T03:28:34.579635Z",
"url": "https://files.pythonhosted.org/packages/98/d7/70264fb178f79f6c7b1981cf6780dc0a2feb4ed76bdea771882b38b971f7/composeml-0.10.1.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-01-07 03:28:34",
"github": false,
"gitlab": false,
"bitbucket": false,
"lcname": "composeml"
}
JSON