feat: optimized r1 pipeline for efficient code RL

Author: ganler

.gitignore

@@ -1,5 +1,6 @@
 # code r1
 data/
+models/
 
 **/*.pt
 **/checkpoints

README-code-r1.md

@@ -1,8 +1,16 @@
-Use the setup script below to avoid environment issues:
+# Code R1
+
+## Setup
 
 ```bash
+# For sandboxing
+sudo add-apt-repository ppa:deki/firejail
+sudo apt-get update
+sudo apt-get install firejail firejail-profiles
+
+# For training
 pip install vllm==0.6.3 torch==2.4.0 ray
-pip3 install flash-attn --no-build-isolation
+pip install flash-attn --no-build-isolation
 pip install -e .  # For verl integration
 pip install wandb IPython matplotlib gpustat
 ```