NCFG-CS_Source_Code

This is the source codes for the NCFG-CS model and other baseline model The Source code is coming soon.

Introduction

This is the deeplearning project for code search.We use Pytorch to implement the model.

Environment dependencies

You can use the following commands to install these dependencies

pip install -r requirements.txt

Dataset

We use the CodeSearchNet dataset, and the preprocessed dataset can be accessed via the cloud drive preprocessed dataset

If you want to quickly start training/testing the model, you can directly obtain ther pocessed data files preprocessed_train/valid/text.bin and place them in the /NCFG-CS_Source_Code/NCFG-CS/preprocessed_data/CSN/train(/valid/test)/processed folder.

models

Our pre-trained models and the corresponding logs are located in /NCFG-CS_Source_Code/NCFG-CS/saved_model.

train model

Use the following commands to train the model

python train.py -batch_size 128 -num_epoch 100 -emb_size 128 -hidden_dim 256 -pool_size 2000 -loss_type cos -saved_model_dir ./saved_model/model1 -train

test model

Use the following commands to test the model

python train.py -batch_size 32 -num_epoch 100 -emb_size 128 -hidden_dim 256 -pool_size 2000 -loss_type cos -pool_size 2000 -test -test_model_path ./saved_model/model1/best_model.bin

baseline models

The rest of the model code can be referred to in the link. baseline code