You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A simple interactive visualisation tool for multi-dimensional data produced with the [PDOPT Code](https://github.com/spinjet/pdopt-code).
### Installation
Just install the latest version of dash plotly using pip:
```
pip install dash
```
### Usage
Run the code from the terminal with the command:
```
$ python dm_tool.py
```
It will output a local IP address which can be opened with a browser.
```
Dash is running on http://127.0.0.1:8050/
* Serving Flask app 'dm_tool_mod'
* Debug mode: off
WARNING: This is a development server. Do not use it in a production deployment. Use a production WSGI server instead.
* Running on http://127.0.0.1:8050
Press CTRL+C to quit
```
Data, in the format of a .csv file, can be loaded by pressing the "Load .csv file" link.
Two types of plots are available: Parallel Coordinates and Scatter.
Parallel Coordinates have a drop-down menu for selecting the axis where the colormap will be applied to.
It is possible to select "None" for providing just plain colors.
Datapoints can be selected by selecting a range of data on each axis. The selections will propagate to the other plots, by greying out the non-selected points.
On the scatter plots it is possible to select using the drop-down menus the X, Y and colour axes.
An example dataset is available in this repository in the example folder.