multipromptify 2.0.1

A tool that creates multi-prompt datasets from single-prompt datasets using templates

MultiPromptify

A tool that creates multi-prompt datasets from single-prompt datasets using templates with variation specifications.

Overview

MultiPromptify transforms your single-prompt datasets into rich multi-prompt datasets by applying various types of variations specified in your templates. It supports HuggingFace-compatible datasets and provides both a command-line interface and a modern web UI.

📚 Documentation

• 📖 Complete API Guide - Python API reference and examples
• 🏗️ Developer Documentation - For contributors and developers
  • Project Structure - Code organization guide
  • Publishing Guide - Package publishing instructions
  • Implementation Summaries - Technical implementation details

Installation

From PyPI (Recommended)

pip install multipromptify

From GitHub (Latest)

pip install git+https://github.com/ehabba/MultiPromptifyPipeline.git

From Source

git clone https://github.com/ehabba/MultiPromptifyPipeline.git
cd MultiPromptifyPipeline
pip install -e .

With Web UI Support

# Install with web UI components
pip install -e ".[ui]"

Quick Start

Web UI (Recommended)

Launch the modern web interface for an intuitive experience:

# From project root
python src/ui/run_streamlit.py

# Or use the demo script
python demo_ui.py

The web UI provides:

• 📁 Step 1: Upload data or use sample datasets
• 🔧 Step 2: Build templates with smart suggestions
• ⚡ Step 3: Generate variations with real-time progress
• 🎉 Step 4: Analyze results and export in multiple formats

Command Line Interface

multipromptify --template "{instruction:semantic}: {col1:paraphrase}" \
               --data data.csv \
               --instruction "Classify the sentiment"

Python API

Using MultiPromptifyAPI (Recommended)

from multipromptify import MultiPromptifyAPI
import pandas as pd

# Initialize
mp = MultiPromptifyAPI()

# Load data
data = [{"question": "What is 2+2?", "answer": "4"}]
mp.load_dataframe(pd.DataFrame(data))

# Configure template
template = {
    'instruction_template': 'Q: {question}\nA: {answer}',
    'question': ['surface'],
    'gold': 'answer'
}
mp.set_template(template)

# Configure and generate
mp.configure(max_rows=1, variations_per_field=3)
variations = mp.generate(verbose=True)

# Export results
mp.export("output.json", format="json")

Using MultiPromptify (Legacy)

from multipromptify import MultiPromptify
import pandas as pd

# Your data
data = pd.DataFrame({
    'question': ['What is 2+2?', 'What color is the sky?'],
    'options': ['A)3 B)4 C)5', 'A)Red B)Blue C)Green']
})

# Template with variation specifications
template = "{instruction:semantic}: {few_shot}\n Question: {question:paraphrase}\n Options: {options}"

# Initialize and generate variations
mp = MultiPromptify()
variations = mp.generate_variations(
    template=template,
    data=data,
    instruction="Choose the correct answer",
    few_shot=["Example: 1+1=2"]
)

print(f"Generated {len(variations)} prompt variations")

Template Format

Templates use Python f-string syntax with custom variation annotations:

"{instruction:semantic}: {few_shot}\n Question: {question:paraphrase}\n Options: {options:non-semantic}"

Supported variation types:

• :semantic - Semantic variations (meaning-preserving)
• :paraphrase - Paraphrasing variations
• :non-semantic - Non-semantic variations (formatting, etc.)
• :lexical - Word choice variations
• :syntactic - Sentence structure variations
• :surface - Surface-level formatting variations

Features

Template System

• Python f-string compatibility: Use familiar {variable} syntax
• Variation annotations: Specify variation types with :type syntax
• Flexible column mapping: Reference any column from your data
• Literal support: Use static strings and numbers

Input Handling

• CSV/DataFrame support: Direct pandas DataFrame or CSV file input
• HuggingFace datasets: Full compatibility with datasets library
• Dictionary inputs: Support for various input types
  • Literals (strings/numbers): Applied to entire dataset
  • Lists: Applied per sample/row
  • Few-shot examples: Flexible list or tuple formats

Web UI Features

• Sample Datasets: Built-in datasets for quick testing
• Template Suggestions: Smart suggestions based on your data
• Real-time Validation: Instant feedback on template syntax
• Live Preview: Test templates before full generation
• Advanced Analytics: Distribution charts, field analysis
• Search & Filter: Find specific variations quickly
• Multiple Export Formats: JSON, CSV, TXT, and custom formats

Few-shot Examples

# Different examples per sample
few_shot = [
    ["Example 1 for sample 1", "Example 2 for sample 1"],
    ["Example 1 for sample 2", "Example 2 for sample 2"]
]

# Same examples for all samples
few_shot = ("Example 1", "Example 2")

Command Line Interface

Basic Commands

# Basic usage
multipromptify --template "{instruction:semantic}: {question:paraphrase}" \
               --data data.csv \
               --instruction "Answer the question"

# With output file
multipromptify --template "{instruction}: {question:paraphrase}" \
               --data data.csv \
               --instruction "Answer this" \
               --output variations.json

# Specify number of variations
multipromptify --template "{instruction:semantic}: {question}" \
               --data data.csv \
               --instruction "Solve this" \
               --max-variations 50

Advanced Options

# With few-shot examples from file
multipromptify --template "{instruction}: {few_shot}\n{question:paraphrase}" \
               --data data.csv \
               --instruction "Answer the question" \
               --few-shot-file examples.txt \
               --few-shot-count 3

# Output to HuggingFace dataset format
multipromptify --template "{instruction:semantic}: {question}" \
               --data data.csv \
               --instruction "Solve this" \
               --output-format hf \
               --output dataset_variations/

API Reference

MultiPromptify Class

class MultiPromptify:
    def __init__(self, max_variations: int = 100):
        """Initialize MultiPromptify generator."""
        
    def generate_variations(
        self,
        template: str,
        data: Union[pd.DataFrame, str, dict],
        instruction: str = None,
        few_shot: Union[list, tuple] = None,
        **kwargs
    ) -> List[Dict[str, Any]]:
        """Generate prompt variations based on template."""
        
    def parse_template(self, template: str) -> Dict[str, str]:
        """Parse template to extract columns and variation types."""
        
    def save_variations(
        self,
        variations: List[Dict[str, Any]],
        output_path: str,
        format: str = "json"
    ):
        """Save variations to file."""

Examples

Sentiment Analysis

import pandas as pd
from multipromptify import MultiPromptify

data = pd.DataFrame({
    'text': ['I love this movie!', 'This book is terrible.'],
    'label': ['positive', 'negative']
})

template = "{instruction:semantic}: '{text:paraphrase}'\nSentiment: {label}"

mp = MultiPromptify()
variations = mp.generate_variations(
    template=template,
    data=data,
    instruction="Classify the sentiment of the following text"
)

Question Answering with Few-shot

template = "{instruction:paraphrase}: {few_shot}\n\nQuestion: {question:semantic}\nAnswer:"

few_shot_examples = [
    "Q: What is the capital of France? A: Paris",
    "Q: What is 2+2? A: 4"
]

variations = mp.generate_variations(
    template=template,
    data=qa_data,
    instruction="Answer the following question",
    few_shot=few_shot_examples
)

Multiple Choice

template = "{instruction:semantic}:\n\n{context:paraphrase}\n\nQuestion: {question}\nOptions:\n{options:non-semantic}\n\nAnswer:"

variations = mp.generate_variations(
    template=template,
    data=mc_data,
    instruction="Choose the best answer"
)

Web UI Screenshots

The MultiPromptify 2.0 web interface provides an intuitive workflow:

1. Data Upload: Upload CSV/JSON files or select from sample datasets
2. Template Builder: Create templates with smart suggestions and real-time validation
3. Generation: Configure settings and watch real-time progress
4. Results: Analyze, search, filter, and export your variations

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests
5. Submit a pull request

License

MIT License - see LICENSE file for details.