Metadata-Version: 2.4
Name: caif-framework
Version: 0.0.1
Summary: CAIF (Code AI Fast): A simple, fast framework for building any AI model.
Author-email: Rudra <rudra@caif.ai>
License: Rudra's Classical Source-Available Licence (RCSAL)
        
        This project (CAIF—Code AI Fast Framework) is licensed under RCSAL — Rudra's Classical Source-Available Licence. It is a source-available licence that allows both commercial and non-commercial use, with conditions outlined below.
        
        ---
        
        TERMS
        
        1. Source Availability
           - The source code of this repository is openly available for inspection, use, and modification.
           - You may use the code in both commercial and non-commercial projects.
        
        2. Editing and Redistribution
           - You are free to edit, modify, and redistribute this code.
           - Condition: Any redistribution (modified or unmodified) must clearly include:
             - The original author’s name (Rudra).
             - A link to the original CAIF Framework repository (to be inserted when uploaded).
             - A clear mention that the project is licensed under RCSAL.
        
        3. Attribution Requirements
           - Attribution is ONLY required when editing or redistributing the CAIF source code.
           - Attribution is NOT required for artifacts generated by the CAIF Framework (e.g., HTML files, Python code snippets, or analytical reports created by the AI).
           - You may freely use, share, and distribute AI-generated code and reports without credit.
        
        4. Commercial Use
           - Commercial use of the CAIF source code is permitted, provided attribution requirements are met.
           - Generated artifacts (HTML, Python Code, etc., authored by the CAIF Framework for end users) may be used commercially without attribution.
        
        5. Non-Commercial Use
           - Non-commercial use is fully permitted under the same conditions.
        
        6. Integrity
           - Do not misrepresent the origin of the project.
           - Any redistributed or modified version must clearly state that it is based on the CAIF Framework and licensed under RCSAL.
           
        7. Anything Else
           - Anything not mentioned is reserved to the owner, Rudra.
        
        ---
        
        SUMMARY
        
        - Free to use (commercial + non-commercial)
        - Free to edit and redistribute, with attribution (name + repo link + licence mention)
        - Generated artifacts (like HTML or Python code) do NOT require credit
        - No misrepresentation of origin
        
        LICENCE NAME
        RCSAL — Rudra's Classical Source-Available Licence
        This licence is not unique to this project (it can be used by others, or me elsewhere) and must be referenced by name in any redistribution.
        
Keywords: AI,machine learning,data,chatbot,nlp,fast-development
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Development Status :: 3 - Alpha
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE.txt
Requires-Dist: numpy
Requires-Dist: pandas
Requires-Dist: scikit-learn
Requires-Dist: Pillow
Requires-Dist: PyYAML
Requires-Dist: openpyxl
Requires-Dist: markdown-it-py
Dynamic: license-file

# ⚡ CAIF (Code AI Fast) Framework

> **Developer:** Rudra  
> **Version:** 0.0.1 (Alpha)

CAIF is a **unified, fast-track AI framework** built on Python. It is designed to significantly reduce the complexity and amount of code needed to train, deploy, and analyze **ANY** type of AI model—from image recognition and data analysis to fully interactive web chatbots.

CAIF handles all the hard parts automatically: data cleaning, intelligent Neural Network sketching, and rich text output generation (Markdown, HTML, & Python Code).

---

## 🚀 Key Features

* **⚡ Code Simplification:** Replaces hundreds of lines of complex Python with simple, one-line commands.
* **🌍 Omnivorous Data Loader:** Natively supports all major file types for training data (MP4, JPG, CSV, EXCEL, JSON, ZIP, etc.) from both local files and URLs.
* **🧠 Auto-Sketched Models:** Automatically designs and optimizes the Neural Network architecture (number of neurons and layers) based on the size and richness of your training data.
* **💬 Rich Text Output (MDP):** Models automatically respond in Markdown, which CAIF instantly converts into clean **HTML** (for websites) or runnable **Python Code** (for rich-text apps).
* **🌐 Specialized Builders:** Includes powerful shortcuts like `CAIF.WEB_CHATBOT()` for rapid, customizable web application creation.

---

## 🛠️ Installation

CAIF is published on PyPI and is easy to install using Python's package installer, `pip`.

```bash
pip install caif-framework
```

## 💡 Quick Start Example

This example shows how to train an AI to predict a value from a simple spreadsheet (CSV) and generate a rich report:
```python
from caif import CAIF

# 1. DATA: Load your data from a URL or file, and specify the column to predict.
# CAIF automatically handles cleaning, formatting, and converting data to "special numbers."
CAIF.DATA(
    source="[https://data.com/house_prices.csv](https://data.com/house_prices.csv)",
    type="csv",
    target_column="Sale_Price"
)

# 2. MODEL: Tell CAIF what kind of AI to build. Complexity (1-10) suggests the size.
# CAIF automatically designs the best Neural Network structure.
CAIF.MODEL(
    type="housing_predictor", 
    complexity=7, 
    time_limit="15m"
)

# 3. ANALYZE (Shortcut): Generates a full performance report of the model.
# The AI generates a Markdown report about its accuracy and key features.
CAIF.ANALYZE(focus="report")

# 4. OUTPUT: Save the final Markdown report as a clean HTML file.
# CAIF's MDP converts the rich text Markdown into HTML code.
CAIF.OUTPUT(target="HTML", save_as="house_price_analysis.html")
```

## 💬 Specialized Web Chatbot Example

Build and style a fully interactive, pirate-themed chatbot interface with one command:
```Python
CAIF.WEB_CHATBOT(
    personality="funny, pirate-themed",
    data_source="C:/MyFiles/pirate_dialogue.jsonl",
    interface="HTML, CSS, JS",
    look_spec="window: wood texture; font: italic gold"
)
```
