CodeGen

Official release for the CodeGen models (350M, 2B, 6B, 16B) for Program Synthesis. That is, the model translates English into executable code as presented in the paper:

Title: A Conversational Paradigm for Program Synthesis

Authors: Erik Nijkamp*, Bo Pang*, Hiroaki Hayashi*, Lifu Tu, Huan Wang, Yingbo Zhou, Silvio Savarese, and Caiming Xiong (* indicates equal contribution)

The current version releases the sampling code, while the detailed training code will be released soon.

HuggingFace

The model is available on the HuggingFace Hub with a Colab demo here.

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("Salesforce/codegen-2B-mono")
model = AutoModelForCausalLM.from_pretrained("Salesforce/codegen-2B-mono")
inputs = tokenizer("# this function prints hello world", return_tensors="pt").to(0)
sample = model.generate(**inputs, max_length=128)
print(tokenizer.decode(sample[0], truncate_before_pattern=[r"\n\n^#", "^'''", "\n\n\n"]))

Colab

This Google Colab notebook allows for sampling from the CodeGen models.

Setup

git clone https://github.com/salesforce/CodeGen
cd CodeGen

# download the model parameters
# codegen-350M-nl,multi,mono
# wget -P checkpoints https://storage.googleapis.com/sfr-codegen-research/checkpoints/codegen-350M-nl.tar.gz && tar -xvf checkpoints/codegen-350M-nl.tar.gz -C checkpoints/
# wget -P checkpoints https://storage.googleapis.com/sfr-codegen-research/checkpoints/codegen-350M-multi.tar.gz && tar -xvf checkpoints/codegen-350M-multi.tar.gz -C checkpoints/
wget -P checkpoints https://storage.googleapis.com/sfr-codegen-research/checkpoints/codegen-350M-mono.tar.gz && tar -xvf checkpoints/codegen-350M-mono.tar.gz -C checkpoints/
# codegen-2B-nl,multi,mono
# wget -P checkpoints https://storage.googleapis.com/sfr-codegen-research/checkpoints/codegen-2B-nl.tar.gz && tar -xvf checkpoints/codegen-2B-nl.tar.gz -C checkpoints/
# wget -P checkpoints https://storage.googleapis.com/sfr-codegen-research/checkpoints/codegen-2B-multi.tar.gz && tar -xvf checkpoints/codegen-2B-multi.tar.gz -C checkpoints/
# wget -P checkpoints https://storage.googleapis.com/sfr-codegen-research/checkpoints/codegen-2B-mono.tar.gz && tar -xvf checkpoints/codegen-2B-mono.tar.gz -C checkpoints/
# codegen-6B-nl,multi,mono
# wget -P checkpoints https://storage.googleapis.com/sfr-codegen-research/checkpoints/codegen-6B-nl.tar.gz && tar -xvf checkpoints/codegen-6B-nl.tar.gz -C checkpoints/
# wget -P checkpoints https://storage.googleapis.com/sfr-codegen-research/checkpoints/codegen-6B-multi.tar.gz && tar -xvf checkpoints/codegen-6B-multi.tar.gz -C checkpoints/
# wget -P checkpoints https://storage.googleapis.com/sfr-codegen-research/checkpoints/codegen-6B-mono.tar.gz && tar -xvf checkpoints/codegen-6B-mono.tar.gz -C checkpoints/
# codegen-16B-nl,multi,mono
# wget -P checkpoints https://storage.googleapis.com/sfr-codegen-research/checkpoints/codegen-16B-nl.tar.gz && tar -xvf checkpoints/codegen-16B-nl.tar.gz -C checkpoints/
# wget -P checkpoints https://storage.googleapis.com/sfr-codegen-research/checkpoints/codegen-16B-multi.tar.gz && tar -xvf checkpoints/codegen-16B-multi.tar.gz -C checkpoints/
# wget -P checkpoints https://storage.googleapis.com/sfr-codegen-research/checkpoints/codegen-16B-mono.tar.gz && tar -xvf checkpoints/codegen-16B-mono.tar.gz -C checkpoints/

# create a virtual environment with requirements
python3.8 -m venv .venv
source .venv/bin/activate
pip3 install --upgrade pip setuptools
pip3 install -r requirements.txt

# sample from the model with an arbitrary context
python3 -m jaxformer.hf.sample --model codegen-350M-mono --context "def hello_world():"

Released Models

We release models of various sizes trained on various datasets. The models are named in the following format:

codegen-{model-size}-{data}

model-size has 4 options: 350M, 2B, 6B, 16B, which represent the number of parameters in each model.

data has 3 options: nl, multi, mono.

nl models are randomly initialized and trained on The Pile, a 825.18 GB English text corpous.
multi models are initialized from nl models and then trained on a corpus with code data consisting of multiple programming languages.
mono models are initialized from multi models and then trained on a corpus with Python code data.

The model names can be provided to the --model flag for sample.py. See a sample usage above in Setup.

Citation

If you find our code or paper useful, please cite the paper:

@article{Nijkamp2022ACP,
  title={A Conversational Paradigm for Program Synthesis},
  author={Nijkamp, Erik and Pang, Bo and Hayashi, Hiroaki and Tu, Lifu and Wang, Huan and Zhou, Yingbo and Savarese, Silvio and Xiong, Caiming},
  journal={arXiv preprint},
  year={2022}
}

License

Our code is BSD-3 licensed. See LICENSE.txt for details.

leslieongit / CodeGen