<div align="center"><img width="100%" src="https://raw.githubusercontent.com/imooger/adix/master/assets/adx1.png"/></div>
---
## _Making Data Science Fun, One Color at a Time!_
# What is it?
**ADIX** is a free, open-source, color-customizable data analysis tool that simplifies Exploratory Data Analysis (EDA) with a single command `ix.eda()`. Experience a streamlined approach to uncovering insights, empowering you to focus on your data without distraction.
**Color customization** is at your fingertips, allowing you to tailor your analysis to your exact needs. Explore your data with confidence and efficiency, knowing that **adix** (Automatic Data Inspection and eXploration) has your back every step of the way.
## ⭐️ if you like the project, please consider giving it a star, thank you :)
# Main Features
- **Customizable Themes**
- Spruce up the **adix** environment with your own personal touch by playing with color schemes!
- **Eficient Cache Utilization**
- Experience faster load times through optimized caching mechanisms, enhancing overall system performance.
- **Rapid Data Insight**
- **adix** prioritizes swiftly showcasing crucial data insights, ensuring quick access to important information.
- **Automatic Type Detection**
- Detects numerical, categorical, and text features automatically, with the option for manual overrides when
necessary.
- **Statistically Rich Summary Information:**
- Unveil the intricate details of your data with a comprehensive summary, encompassing type identification, unique values, missing values, duplicate rows, the most frequent values and more.
- Delve deeper into numerical data, exploring properties like min-max range, quartiles, average, median, standard deviation, variance, sum, kurtosis, skewness and more.
- **Univariate and Bivariate Statistics Unveiled**
- Explore univariate and bivariate insights with adix's versatile visualization options. From bar charts to matrices, and box plots, uncover a multitude of ways to interpret and analyze your data effectively.
# Documentation
[Docs](https://imooger.github.io)
# Installation
The best way to install **adix** (other than from source) is to use pip:
```
pip install adix
```
**adix is still under development** If you encounter any data, compatibility, or installation issues, please don't hesitate to reach out!
# Quick start
The system is designed for rapid visualization of target values and dataset, facilitating quick analysis of target characteristics with just one function `ix.eda()`. Similar to pandas' df.describe() function, it provides extended analysis capabilities, accommodating time-series and text data for comprehensive insights.
```python
import adix as ix
from adix.datasets load_dataset
titanic = load_dataset('titanic')
```
## 10 minutes to **adix**
### 1. Rendering the whole dataframe
```python
ix.eda(titanic)
```
- using _forest color theme_
---
### 2. Accesing variables of specific dtype
Render the DataFrame containing only categorical variables.
```python
ix.eda(titanic,vars='categorical')
```
---
### 3. Accesing individual variables
```python
ix.eda(titanic,'Age')
```
- using _forest color theme_
---
### 4. Pandas .loc & .iloc
An easy way to render only a part of the DataFrame you are interested in.
```python
ix.eda(titanic.loc[:10:2,['Age','Pclass','Fare'])
```
---
### 5. Changing theme colors
```python
ix.Configs.get_theme()
...
ix.Configs.set_theme('FOREST')
```
<div align="center"><img width="100%" src="https://raw.githubusercontent.com/imooger/adix/master/assets/change_c.png"/></div>
---
### 6. Heatmap correlation
This visualization depicts the correlation between all numerical variables within the DataFrame, offering valuable insights into the magnitude and direction of their relationships.
```python
# Show correlation for the entire DataFrame.
ix.eda(titanic,corr=True)
```
<div align="center"><img width="100%" src="https://raw.githubusercontent.com/imooger/adix/master/assets/corr_main.png"/></div>
Furthermore, it is possible to use categorical variables since they undergo one-hot encoding to enable their inclusion in correlation analysis. It's recommended to use ANOVA. You can choose whatever variables you want to explore and analyze.
```python
# Show correlation for selected parts of the DataFrame
ix.eda(titanic.loc[:,['Age','Fare','Sex','Survived']],vars=['categorical','continuous'],corr=True)
```
<div align="center"><img width="100%" src="https://raw.githubusercontent.com/imooger/adix/master/assets/corr_small.png"/></div>
---
### 7. Bivariate relationships: numerical & numerical
```python
ix.eda(titanic,'Age','Fare')
```
<div align="center"><img width="100%" src="https://raw.githubusercontent.com/imooger/adix/master/assets/c_c.png"/></div>
---
### 8. Bivariate relationships: categorical & numerical
```python
ix.eda(titanic,'Sex','Age')
```
<div align="center"><img width="100%" src="https://raw.githubusercontent.com/imooger/adix/master/assets/cat_c.png"/></div>
---
### 9. Bivariate relationships: categorical & categorical
```python
ix.eda(titanic,'Sex','Survived')
```
<div align="center"><img width="100%" src="https://raw.githubusercontent.com/imooger/adix/master/assets/cat_cat.png"/></div>
## License
MIT
**Free Software, Hell Yeah!**
## Development
Contributions are welcome, so feel free to contact, open an issue, or submit a pull request!
For accessing the codebase or reporting bugs, please visit the GitHub repository.
This program is provided WITHOUT ANY WARRANTY. **ADIX** is still under heavy development and there might be hidden bugs.
## Acknowledgement
The goal for **adix** is to make valuable information and visualization readily available in a user friendly environment at the click of a mouse, without reinventing the wheel. All of the libraries stated below are powerful and excellent alternatives to adix. Several functions of **adix** were inspired from the following:
- [Sweetviz](https://github.com/fbdesignpro/sweetviz) : The inception of this project found inspiration from Sweetviz, particularly its concept of consolidating all data in one place and using the blocks for individual features.
- [Dataprep](https://github.com/sfu-db/dataprep/tree/develop) : Dataprep stands out as an excellent library for data preparation, and certain structural elements of adix have been inspired by it.
- [Pandas-Profiling](https://github.com/ydataai/ydata-profiling) : Alerts served as inspiration for a segment of the dashboard's design, contributing to its functionality and user-friendly features."
- [Kaggle](https://kaggle.com) source of Titanic dataset
Raw data
{
"_id": null,
"home_page": "",
"name": "adix",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.10",
"maintainer_email": "",
"keywords": "pandas eda data data-science data-analysis python jupyter ipython",
"author": "Marek Jindrich",
"author_email": "adix.eda@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/a8/94/0d9906a7d64ac2c93d519f09be579b5e9b224278d79d8efe972ed558beb4/adix-0.2.6.tar.gz",
"platform": null,
"description": "<div align=\"center\"><img width=\"100%\" src=\"https://raw.githubusercontent.com/imooger/adix/master/assets/adx1.png\"/></div>\n\n---\n\n## _Making Data Science Fun, One Color at a Time!_\n\n\n\n\n\n\n\n\n# What is it?\n**ADIX** is a free, open-source, color-customizable data analysis tool that simplifies Exploratory Data Analysis (EDA) with a single command `ix.eda()`. Experience a streamlined approach to uncovering insights, empowering you to focus on your data without distraction. \n**Color customization** is at your fingertips, allowing you to tailor your analysis to your exact needs. Explore your data with confidence and efficiency, knowing that **adix** (Automatic Data Inspection and eXploration) has your back every step of the way.\n\n\n\n\n## \u2b50\ufe0f if you like the project, please consider giving it a star, thank you :)\n\n# Main Features\n- **Customizable Themes** \n - Spruce up the **adix** environment with your own personal touch by playing with color schemes! \n- **Eficient Cache Utilization**\n - Experience faster load times through optimized caching mechanisms, enhancing overall system performance. \n- **Rapid Data Insight**\n - **adix** prioritizes swiftly showcasing crucial data insights, ensuring quick access to important information. \n- **Automatic Type Detection**\n - Detects numerical, categorical, and text features automatically, with the option for manual overrides when\n necessary.\n- **Statistically Rich Summary Information:** \n - Unveil the intricate details of your data with a comprehensive summary, encompassing type identification, unique values, missing values, duplicate rows, the most frequent values and more. \n - Delve deeper into numerical data, exploring properties like min-max range, quartiles, average, median, standard deviation, variance, sum, kurtosis, skewness and more.\n- **Univariate and Bivariate Statistics Unveiled**\n - Explore univariate and bivariate insights with adix's versatile visualization options. From bar charts to matrices, and box plots, uncover a multitude of ways to interpret and analyze your data effectively.\n \n# Documentation\n[Docs](https://imooger.github.io)\n\n# Installation\nThe best way to install **adix** (other than from source) is to use pip:\n```\npip install adix\n```\n\n**adix is still under development** If you encounter any data, compatibility, or installation issues, please don't hesitate to reach out!\n\n\n# Quick start\nThe system is designed for rapid visualization of target values and dataset, facilitating quick analysis of target characteristics with just one function `ix.eda()`. Similar to pandas' df.describe() function, it provides extended analysis capabilities, accommodating time-series and text data for comprehensive insights.\n\n```python\nimport adix as ix\nfrom adix.datasets load_dataset\n\ntitanic = load_dataset('titanic')\n```\n\n## 10 minutes to **adix**\n\n\n### 1. Rendering the whole dataframe\n```python\nix.eda(titanic)\n```\n- using _forest color theme_\n\n\n\n---\n### 2. Accesing variables of specific dtype\nRender the DataFrame containing only categorical variables.\n\n```python\nix.eda(titanic,vars='categorical')\n```\n---\n### 3. Accesing individual variables\n```python\nix.eda(titanic,'Age')\n```\n- using _forest color theme_\n \n\n\n---\n### 4. Pandas .loc & .iloc\nAn easy way to render only a part of the DataFrame you are interested in.\n\n```python\nix.eda(titanic.loc[:10:2,['Age','Pclass','Fare'])\n```\n---\n\n### 5. Changing theme colors\n```python\nix.Configs.get_theme()\n...\nix.Configs.set_theme('FOREST')\n```\n<div align=\"center\"><img width=\"100%\" src=\"https://raw.githubusercontent.com/imooger/adix/master/assets/change_c.png\"/></div>\n\n---\n\n\n### 6. Heatmap correlation\nThis visualization depicts the correlation between all numerical variables within the DataFrame, offering valuable insights into the magnitude and direction of their relationships. \n\n\n```python\n# Show correlation for the entire DataFrame.\nix.eda(titanic,corr=True)\n\n```\n<div align=\"center\"><img width=\"100%\" src=\"https://raw.githubusercontent.com/imooger/adix/master/assets/corr_main.png\"/></div>\n\n\nFurthermore, it is possible to use categorical variables since they undergo one-hot encoding to enable their inclusion in correlation analysis. It's recommended to use ANOVA. You can choose whatever variables you want to explore and analyze.\n\n\n```python\n# Show correlation for selected parts of the DataFrame\nix.eda(titanic.loc[:,['Age','Fare','Sex','Survived']],vars=['categorical','continuous'],corr=True)\n\n```\n<div align=\"center\"><img width=\"100%\" src=\"https://raw.githubusercontent.com/imooger/adix/master/assets/corr_small.png\"/></div>\n\n\n---\n\n### 7. Bivariate relationships: numerical & numerical\n```python\nix.eda(titanic,'Age','Fare')\n\n```\n<div align=\"center\"><img width=\"100%\" src=\"https://raw.githubusercontent.com/imooger/adix/master/assets/c_c.png\"/></div>\n\n---\n\n### 8. Bivariate relationships: categorical & numerical\n```python\nix.eda(titanic,'Sex','Age')\n\n\n```\n<div align=\"center\"><img width=\"100%\" src=\"https://raw.githubusercontent.com/imooger/adix/master/assets/cat_c.png\"/></div>\n\n---\n\n### 9. Bivariate relationships: categorical & categorical\n```python\nix.eda(titanic,'Sex','Survived')\n\n```\n<div align=\"center\"><img width=\"100%\" src=\"https://raw.githubusercontent.com/imooger/adix/master/assets/cat_cat.png\"/></div>\n\n\n\n\n\n## License\n\nMIT\n\n**Free Software, Hell Yeah!**\n\n\n## Development\n\nContributions are welcome, so feel free to contact, open an issue, or submit a pull request!\n\nFor accessing the codebase or reporting bugs, please visit the GitHub repository.\n\nThis program is provided WITHOUT ANY WARRANTY. **ADIX** is still under heavy development and there might be hidden bugs.\n\n\n## Acknowledgement\n\nThe goal for **adix** is to make valuable information and visualization readily available in a user friendly environment at the click of a mouse, without reinventing the wheel. All of the libraries stated below are powerful and excellent alternatives to adix. Several functions of **adix** were inspired from the following:\n\n- [Sweetviz](https://github.com/fbdesignpro/sweetviz) : The inception of this project found inspiration from Sweetviz, particularly its concept of consolidating all data in one place and using the blocks for individual features.\n- [Dataprep](https://github.com/sfu-db/dataprep/tree/develop) : Dataprep stands out as an excellent library for data preparation, and certain structural elements of adix have been inspired by it.\n- [Pandas-Profiling](https://github.com/ydataai/ydata-profiling) : Alerts served as inspiration for a segment of the dashboard's design, contributing to its functionality and user-friendly features.\"\n- [Kaggle](https://kaggle.com) source of Titanic dataset\n\n\n",
"bugtrack_url": null,
"license": "",
"summary": "Automated exploratory data analysis (EDA).",
"version": "0.2.6",
"project_urls": {
"Bug Tracker": "https://github.com/imooger/adix/issues",
"Documentation": "https://imooger.github.io",
"Homepage": "https://imooger.github.io",
"Repository": "https://github.com/imooger/adix"
},
"split_keywords": [
"pandas",
"eda",
"data",
"data-science",
"data-analysis",
"python",
"jupyter",
"ipython"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "9102c5ed60b6deece2619e98746491a037bffcee2b8b3e9d42b3c1600926d45c",
"md5": "a884217d34cdd8461498aec441b1e907",
"sha256": "30ba307cb2c0eaf4b3fb994c799d7f6eca72c08cc7e2e218ab2d1cc5c05e1cf6"
},
"downloads": -1,
"filename": "adix-0.2.6-py3-none-any.whl",
"has_sig": false,
"md5_digest": "a884217d34cdd8461498aec441b1e907",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.10",
"size": 53115,
"upload_time": "2024-02-20T05:44:23",
"upload_time_iso_8601": "2024-02-20T05:44:23.768917Z",
"url": "https://files.pythonhosted.org/packages/91/02/c5ed60b6deece2619e98746491a037bffcee2b8b3e9d42b3c1600926d45c/adix-0.2.6-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "a8940d9906a7d64ac2c93d519f09be579b5e9b224278d79d8efe972ed558beb4",
"md5": "825bade0c60ecfbd9681759de52ac29a",
"sha256": "34b72e1b51278f664e50651a8128c7a49579d194410b069ae14ff90eefcfb235"
},
"downloads": -1,
"filename": "adix-0.2.6.tar.gz",
"has_sig": false,
"md5_digest": "825bade0c60ecfbd9681759de52ac29a",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.10",
"size": 55317,
"upload_time": "2024-02-20T05:44:27",
"upload_time_iso_8601": "2024-02-20T05:44:27.211879Z",
"url": "https://files.pythonhosted.org/packages/a8/94/0d9906a7d64ac2c93d519f09be579b5e9b224278d79d8efe972ed558beb4/adix-0.2.6.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-02-20 05:44:27",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "imooger",
"github_project": "adix",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"lcname": "adix"
}
JSON
