clusterfun 0.1.1a6

Clusterfun - a plotting library to inspect data

Clusterfun

Clusterfun is a tool for visualizing data and media in a browser. It's built on top of Plotly. The goal is to make it easy to visualize data and media in a browser, without having to write much code.

Usage

Plot types

The following plot types are available:

• Scatter
• Histogram
• Grid
• Violin
• Pie chart

import clusterfun as clt

Scatter

clt.scatter(df, x="x", y="y", media="img_path")

Example

Histogram

clt.histogram(df, x="x", media="img_path", bins=50)

Example

Grid

A simple grid of media, no plot. Useful for inspecting data when no underlying numbers are available yet.

clt.grid(df, media="img_path")

Example

Violin

clt.violin(df, y="y", media="img_path")

Example

Pie chart

clt.pie(df, x="x", y="y", media="img_path")

Example

Parameters

Media

The media column in the dataframe will be used to load the media.

Color

You can color different categories with the color parameter.

• The color can be either a color name or hex value

Bounding box

Bounding boxes can be added with the bounding_box parameter. A bounding box cell in a dataframe needs to contain a dictionary or a list of dictionaries with bounding box values: xmin, ymin, xmax, ymax, label (optional), color (optional).
Example of a bounding box:

bounding_box = {
    "xmin": 12,
    "ymin": 10,
    "xmax": 100,
    "ymax": 110,
    "color": "green",
    "label": "ground truth"
}

• The bounding box coordinates can be either floats or integers.

Filtering

Data in plots can be filtered to quickly find subsets of the data you're interested in.

Installation

Python library

You can create plots that open in your browser by installing the Python library:

pip install clusterfun

Data loading

Clusterfun supports S3 and local data storage and loading. The media column in the dataframe will be used to determine where to load the media from.