Metadata-Version: 2.1
Name: sensor-dataset
Version: 0.0.1
Summary: Put a description
Home-page: https://github.com/kikejimenez/sensor_dataset/tree/master/
Author: Enrique Jimenez
Author-email: physieira@gmail.com
License: Apache Software License 2.0
Keywords: outlier detection statistics
Platform: UNKNOWN
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: Natural Language :: English
Classifier: Programming Language :: Python :: 3.6
Classifier: Programming Language :: Python :: 3.7
Classifier: Programming Language :: Python :: 3.8
Classifier: License :: OSI Approved :: Apache Software License
Requires-Python: >=3.6
Description-Content-Type: text/markdown
Requires-Dist: pip
Requires-Dist: packaging
Requires-Dist: pandas
Requires-Dist: plotly
Requires-Dist: scipy
Requires-Dist: koalas
Requires-Dist: pyspark
Requires-Dist: py4j
Requires-Dist: kaleido
Provides-Extra: dev

# Outlier Detection
> Detect and filter outliers.


> [Documentation and Code can be found on Github]()

## Install

`pip install sensor_dataset`

## Z-SCORE Normalization

> Normalize data with Z-SCORE

```python
from sensor_dataset.outlier_detection import ZSCORE
```

Get a normalized Koalas dataframe for the sensor dataset and fig objects by calling:

```python
kdf, figs = ZSCORE()

figs['NORMAL'].write_image("images/zscore_normal.png")
figs['RECOVERING'].write_image("images/zscore_recovering.png")
figs['BROKEN'].write_image("images/zscore_broken.png")
```

<img src="https://raw.githubusercontent.com/kikejimenez/sensor_dataset/master/nbs/images/zscore_normal.png" width="400" height="300" style="max-width: 400px">
<img src="https://raw.githubusercontent.com/kikejimenez/sensor_dataset/master/nbs/images/zscore_recovering.png" width="400" height="300" style="max-width: 400px">
<img src="https://raw.githubusercontent.com/kikejimenez/sensor_dataset/master/nbs/images/zscore_broken.png" width="400" height="300" style="max-width: 400px">

When running on a notebook you may show an interactive plot by using:
```python
fig.show()
```

## IQR

> Filter data using IQR

```python
from sensor_dataset.outlier_detection import IQR

kdf, figs = IQR()

figs['NORMAL'].write_image("images/iqr_normal.png")
figs['RECOVERING'].write_image("images/iqr_recovering.png")
figs['BROKEN'].write_image("images/iqr_broken.png")
```

<img src="https://raw.githubusercontent.com/kikejimenez/sensor_dataset/master/nbs/images/iqr_normal.png" width="400" height="300" style="max-width: 400px">
<img src="https://raw.githubusercontent.com/kikejimenez/sensor_dataset/master/nbs/images/iqr_recovering.png" width="400" height="300" style="max-width: 400px">
<img src="https://raw.githubusercontent.com/kikejimenez/sensor_dataset/master/nbs/images/iqr_broken.png" width="400" height="300" style="max-width: 400px">


