A python program that visualizes data from a .csv
file. It uses the PySimpleGui python library for its GUI.
At the moment, this program can only read data from .csv
files, but in the future I plan on adding support for other filetypes, and an option to connect to a SQL or a MongoDb database.
If you want to test the program yourself, I recommend you visit the page mockaroo to get some sample data in .csv
format.
Data visualizer can be installed with pip, along with all the program dependencies in the case they aren't installed. To do so, run the follow command:
pip install data-visualizer
On the other hand, if you want to, you can also download the program from the repo. If you want to do so, follow these steps to do so.
- Clone the repo into your local machine:
- Open the cloned folder and run the
"data_visualizer"
folder as a module with the following command:
python3 -m data_visualizer
There are several ways in which you can open the program.
If Data Visualizer was installed with pip, you can run the program in the terminal by simple writing data-visualizer
in the terminal. The program will then open.
Another way you can run the program is by opening either the module folder installed by pip, or the cloned folder from the repo. Then, once inside the folder, simply run the data_visualizer
folder as a python module with the following command:
python3 -m data_visualizer
By default, Data Visualizer's setup.py
file was setup in a way that automatically downloads all the packages the program uses and depends of. However, if for some reason you want to install them yourself, Data Visualizer requires the PySimpleGui library to work.
You can install them with pip using the following commands:
pip install PySimpleGUI
In the case that you don't have pip, scroll down to the Pip Instalation section to see how to install it.
Data Visualizer is made using the PySimpleGui library, which uses Tkinter under the hood. So, it is very important to have Tkinter already setup to your Python version for Data Visualizer to work.
If you don't have Tkinter already with your Python configuration, you will need to do that first.
- Windows: Follow this video guide.
- MacOs (Using Homebrew):
brew install python-tk
- Linux (Ubuntu):
sudo apt-get update && sudo apt-get install python3-tk
Don't have pip? You can install it like this:
python3 -m ensurepip
py -m ensurepip
Please note that the installation of pip in windows can sometimes be... complicated. If you are having problems making pip work, I recommend you this video guide on how to install it.
Data visualizer supports the following types of files:
File Type | Support |
---|---|
.csv | ✅ |
.json | ❌ |
SQL | 📅 |
Mongodb | ❌ |
Keywords: ✅ Yes
, ❌ No
, 📅 work in progress
Currently, there's a bug where the first row of header-less .csv
files gets confused for the file's headers. This happens because Data Visualizer always takes the first row of a .csv file that comes with pre-defined headers, and uses it as the headers of the file. The usefulness of this feature can be seen with files that has its data pre-categorize in the forms of these headers. However, for files that don't do this , this feature becomes a problem.
The nature of this bug makes it pretty difficult to fix, but with time, I think I'll be able to get it done. So, keep that in mind while working with files wihout a proper header for the time being.
Note: The files used in this example were taken from mockaroo.
This project makes use of the MIT license. To read more about it, you can check it out here.
This project makes uses of the PySimpleGui python library, which makes use of the LGPL-3.0 license. Read more about this license in the project repo here.