Pseudocode-To-C++ and C++-To-Pseudocode Converter Using Transformers

A bidirectional converter that translates between C++ code and pseudocode using transformer-based neural networks. The project implements both code-to-pseudocode and pseudocode-to-code translation capabilities using PyTorch.

Features

• Bidirectional translation between C++ code and pseudocode
• Interactive web interface using Streamlit
• Trained on a comprehensive dataset of code-pseudocode pairs
• Support for common programming constructs:
  • Control structures (if-else, loops)
  • Function declarations
  • Input/output operations
  • Basic arithmetic and logical operations

Project Structure

├── app.py                  # Streamlit web application
├── requirements.txt        # Project dependencies
├── spoc_train.csv         # Training dataset
├── code_to_pseudocode/    # Code to Pseudocode translation model
│   ├── model.py
│   ├── code_tokenizer.model
│   ├── text_tokenizer.model
│   └── code_psd_transformer.pt
├── pseudocode_to_code/    # Pseudocode to Code translation model
│   ├── model.py
│   ├── code_tokenizer.model
│   ├── text_tokenizer.model
│   └── psd_code_transformer.pt
└── Notebooks/             # Development and training notebooks
    ├── Code to PseudoCode.ipynb
    └── PseudoCode to Code.ipynb

Technical Details

• Architecture: Transformer-based sequence-to-sequence model
• Components:
  • Multi-head attention mechanism
  • Position-wise feed-forward networks
  • Residual connections and layer normalization
  • SentencePiece tokenization

Installation

1. Clone the repository:

git clone <https://github.com/saadsohail05/Pseudocode-To-Cpp-And-Cpp-To-Pseudocode-Converter-Using-Transformers-url>
cd Pseudocode-To-Cpp-And-Cpp-To-Pseudocode-Converter-Using-Transformers

2. Install dependencies:

pip install -r requirements.txt

Usage

1. Start the web application:

streamlit run app.py

2. Use the interactive interface to:
   • Convert C++ code to pseudocode
   • Convert pseudocode to C++ code
   • View example translations

Model Training

The models were trained on a dataset of C++ code and corresponding pseudocode pairs. The training process involved:

• Data preprocessing and tokenization using SentencePiece
• Training separate transformer models for each direction
• Optimization using AdamW optimizer with cosine annealing
• Validation using BLEU score metrics

Dependencies

• PyTorch
• Streamlit
• SentencePiece
• NumPy
• pandas
• tqdm
• sacrebleu

Examples

C++ to Pseudocode:

for(int i=0; i<n; i++) { 
    sum += arr[i]; 
}

↓ Converts to ↓

loop from 0 to n-1:
    add arr[i] to sum

Pseudocode to C++:

if x is greater than y then
    set max to x
else
    set max to y

↓ Converts to ↓

if(x > y) {
    max = x;
} else {
    max = y;
}

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.