Open-source code generation model code_trans_t5_base_program_synthese - Convert natural language to program code for free

Code Trans T5 Base Program Synthese

Developed by SEBIS

Lisp-style DSL code generation model based on T5 architecture for converting natural language descriptions into program code

Large Language Model #Lisp code generation #Program synthesis #Natural language to DSL

Downloads 16

Release Time : 3/2/2022

Model Overview

This model is specifically designed for program synthesis tasks, capable of generating Lisp-style Domain-Specific Language (DSL) code from natural language descriptions. Pre-trained on the t5-base architecture using SentencePiece tokenizer.

Model Features

Lisp-style DSL generation

Optimized specifically for Lisp-style Domain-Specific Languages, accurately understanding natural language descriptions and converting them into corresponding code

Based on T5 architecture

Utilizes the powerful t5-base model architecture with excellent sequence-to-sequence conversion capabilities

Specialized for program synthesis

Specifically trained for program synthesis tasks, demonstrating outstanding performance in code generation

Model Capabilities

Natural language understanding

Code generation

Program synthesis

Lisp-style DSL conversion

Use Cases

Programming assistance

Algorithm implementation

Generate Lisp-style implementation code from natural language descriptions of algorithm logic

BLEU score up to 90.31

Mathematical operation conversion

Convert mathematical operation descriptions into Lisp-style DSL code

Education

Programming education

Help students understand the relationship between natural language descriptions and code implementation

🚀 CodeTrans model for program synthesis

This is a pre - trained model for program synthesis. It's built on a programming language Lisp - inspired DSL, leveraging the T5 base model architecture. It was first released in this repository.

🚀 Quick Start

The CodeTrans model can generate Lisp - inspired DSL code based on human - language descriptions. Here's a quick guide on how to use it.

✨ Features

Based on the t5 - base model.
Has its own SentencePiece vocabulary model.
Utilizes single - task training on the Program Synthesis dataset.

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

Here is how to use this model to generate lisp inspired DSL code using Transformers SummarizationPipeline:

from transformers import AutoTokenizer, AutoModelWithLMHead, SummarizationPipeline

pipeline = SummarizationPipeline(
    model=AutoModelWithLMHead.from_pretrained("SEBIS/code_trans_t5_base_program_synthese"),
    tokenizer=AutoTokenizer.from_pretrained("SEBIS/code_trans_t5_base_program_synthese", skip_special_tokens=True),
    device=0
)

tokenized_code = "you are given an array of numbers a and a number b , compute the difference of elements in a and b"
pipeline([tokenized_code])

Run this example in colab notebook.

📚 Documentation

Model description

This CodeTrans model is based on the t5 - base model. It has its own SentencePiece vocabulary model. It used single - task training on Program Synthesis dataset.

Intended uses & limitations

The model could be used to generate lisp inspired DSL code based on the human language description tasks.

🔧 Technical Details

Training data

The supervised training tasks datasets can be downloaded on Link

Evaluation results

For the code documentation tasks, different models achieve the following results on different programming languages (in BLEU score):

Test results :

Language / Model	LISP
CodeTrans - ST - Small	89.43
CodeTrans - ST - Base	89.65
CodeTrans - TF - Small	90.30
CodeTrans - TF - Base	90.24
CodeTrans - TF - Large	90.21
CodeTrans - MT - Small	82.88
CodeTrans - MT - Base	86.99
CodeTrans - MT - Large	90.27
CodeTrans - MT - TF - Small	90.31
CodeTrans - MT - TF - Base	90.30
CodeTrans - MT - TF - Large	90.17
State of the art	85.80

📄 License

No license information is provided in the original document.

Created by Ahmed Elnaggar | LinkedIn and Wei Ding | LinkedIn

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご