Metadata-Version: 2.1
Name: clusterysis
Version: 1.0.3
Summary: A library for visualizing clusters.
Home-page: https://github.com/gabrieldi95/clusterysis
Author: Gabriel Di Pardi Arruda
License: UNKNOWN
Platform: UNKNOWN
Description-Content-Type: text/markdown

# Clusterysis

Clusterysis is a simple tool for a first cluster analysis that helps you visualize the difference between clusters.

## Usage:

`import clusterysis as cly`

Just pass a Dataframe with a cluster column to see several charts.

`cly.View(df=df, cluster_col="cluster")`

A table with the mean of each quantitative variable:

![](https://raw.githubusercontent.com/gabrieldi95/clusterysis/master/doc/img/table.png)

Cluster composition of each categorical variable:

![](https://raw.githubusercontent.com/gabrieldi95/clusterysis/master/doc/img/pie.png)

Boxplot chart for quantitative variables:

![](https://raw.githubusercontent.com/gabrieldi95/clusterysis/master/doc/img/box.png)

## Parameters:

`df` and `cluster_col` are required, if you want you can explicit which are the quantitative and categorical variables (if you don't it will be assumed strings are categorical and numbers quantitative).
You can also ask to highlight a specif cluster.

```
cly.View(
    df=df,
    cluster_col="cluster",
    cat_cols=["job", "education"],
    quant_cols=["age", "duration"],
    diff_cluster=0
)
```
![](https://raw.githubusercontent.com/gabrieldi95/clusterysis/master/doc/img/box2.png)

