Data_explorer
Description
This is a project designed to meet the requirements outlined at the bottom of the page. In short, the project aims to facilitate the analysis of CSV files.
Source: Python/Data_Explorer
My Contributions at Work
As part of the work division, I was mainly assigned to backend tasks. There, I created appropriate routes that enabled operations such as file saving, setting the mean, updating files and columns, etc. Occasionally, I also made some small changes in the frontend, such as adding a button or improving the display of empty columns and rows in a different color.
App appearance
On the main screen, you can see the loaded CSV file, and you can click on a column, save the entire file, or just individual columns.
On the column screen, we see all parameters such as mean, median, etc. We can fill in empty cells or manually change them using the update option.
We can also compare two columns, where we get the most important information about them in one place.
Requirements
Project: Application for preliminary data analysis and visualization. Assumptions: Single-stage login. Data is retrieved from CSV files. Number of variables (up to 100). Expected functionalities:
- Automatic detection of data types with editing capability.
- Simple variable name changes.
- Basic 1D EDA descriptive statistics (appropriate to variable type) - expected suggestions.
- Basic visualizations of a single variable (appropriate to variable type) within a specified time range - expected suggestions.
- Filling in missing values / correcting outlier values for variables (appropriate to variable type) - expected suggestions (ability to easily return to point 3).
- Basic measures of dependencies between variables 2D EDA (appropriate to variable types) - expected suggestions.
- Basic visualizations of two/three variables (appropriate to variable types) within a specified time range - expected suggestions. Additionally:
- data retrieval by the user (CSV file) from a specified range (variable selection, case selection).
- ability to save generated plots in jpg and png formats. The output of the application is a “cleaned” data set for further analysis.