Code Safety

...

---

🛠️ Full Setup

1. Environment Setup

git clone https://github.com/FSoft-AI4Code/SafetyCode.git && cd SafetyCode
conda create -n verl_safe python=3.10 -y
conda activate verl_safe
bash install.sh

2. Training

Run training for your target dataset and model size:

# For SecCodePLT+  
bash run/seccodeplt/hybrid-[0.5b|3b|7b].sh

# For APPS+  
bash run/apps-code/hybrid-[0.5b|3b|7b].sh

# For SafeSQL  
bash run/safesql/hybrid-[0.5b|3b|7b].sh

💾 Checkpoints are automatically saved to: checkpoints/[proj_name]/[exp_name]

3. Evaluation

Execute the evaluation pipeline after training:

# For SecCodePLT+
bash experiments/seccodeplt/seccodeplt_eval_main.sh

# For APPS+
bash experiments/safesql_and_apps/apps_eval_main.sh [path_to_exp_dir] [sharded|single]

# For SafeSQL
bash experiments/safesql_and_apps/safesql_eval_main.sh [path_to_exp_dir] [sharded|single]

This codebase is inspired by this paper. Thanks to the authors for their great work.

---