[WIP] VeOmni RL Engine

scripts/install_veomni.sh

@@ -0,0 +1,48 @@
+#!/bin/bash
+
+
+
+export MAX_JOBS=32
+eval "$(conda shell.bash hook)"
+conda create -n verlVeomni python=3.11 -y
+conda activate verlVeomni
+
+python -m pip install torch==2.8.0 torchvision==0.23.0 torchaudio==2.8.0 --index-url https://download.pytorch.org/whl/cu128
+uv pip install --no-cache-dir "vllm==0.11.0"
+
+echo "2. install basic packages"
+uv pip install "transformers[hf_xet]>=4.51.0" accelerate datasets peft hf-transfer \
+    "numpy<2.0.0" "pyarrow>=15.0.0" pandas "tensordict>=0.8.0,<=0.10.0,!=0.9.0" torchdata \
+    ray[default] codetiming hydra-core pylatexenc qwen-vl-utils wandb dill pybind11 liger-kernel mathruler \
+    pytest py-spy pre-commit ruff tensorboard 
+
+echo "pyext is lack of maintainace and cannot work with python 3.12."
+echo "if you need it for prime code rewarding, please install using patched fork:"
+echo "pip install git+https://github.com/ShaohonChen/PyExt.git@py311support"
+
+uv pip install "nvidia-ml-py>=12.560.30" "fastapi[standard]>=0.115.0" "optree>=0.13.0" "pydantic>=2.9" "grpcio>=1.62.1"
+
+
+echo "3. install FlashAttention and FlashInfer"
+# FlashAttention will be installed automatically by VeOmni[gpu]
+# The correct version (2.8.3+cu12torch2.8cxx11abiTRUE) is specified in VeOmni's pyproject.toml
+
+# pip install --no-cache-dir flashinfer-python==0.3.1
+
+
+
+echo "5. May need to fix opencv"
+uv pip install opencv-python
+uv pip install opencv-fixer && \
+    python -c "from opencv_fixer import AutoFix; AutoFix()"
+
+
+echo "6. Install veomni"
+
+git clone https://github.com/ByteDance-Seed/VeOmni.git
+cd VeOmni
+uv pip install -e .[gpu]
+
+
+
+echo "Successfully installed all packages"