Merge branch 'verl-project:main' into main #4
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # # Tests layout | |
| # Each folder under tests/ corresponds to a test category for a sub-namespace in verl. For instance: | |
| # - `tests/trainer` for testing functionality related to `verl/trainer` | |
| # - `tests/models` for testing functionality related to `verl/models` | |
| # - ... | |
| # There are a few folders with `special_` prefix, created for special purposes: | |
| # - `special_distributed`: unit tests that must run with multiple GPUs | |
| # - `special_e2e`: end-to-end tests with training/generation scripts | |
| # - `special_npu`: tests for NPUs | |
| # - `special_sanity`: a suite of quick sanity tests | |
| # - `special_standalone`: a set of test that are designed to run in dedicated environments | |
| # Accelerators for tests | |
| # - By default tests are run with GPU available, except for the ones under `special_npu`, and any test script whose name ends with `on_cpu.py`. | |
| # - For test scripts with `on_cpu.py` name suffix would be tested on CPU resources in linux environment. | |
| # # Workflow layout | |
| # All CI tests are configured by yaml files in `.github/workflows/`. Here's an overview of all test configs: | |
| # 1. A list of always triggered CPU sanity tests: `check-pr-title.yml`, `secrets_scan.yml`, `check-pr-title,yml`, `pre-commit.yml`, `doc.yml` | |
| # 2. Some heavy multi-GPU unit tests, such as `model.yml`, `vllm.yml`, `sgl.yml` | |
| # 3. End-to-end tests: `e2e_*.yml` | |
| # 4. Unit tests | |
| # - `cpu_unit_tests.yml`, run pytest on all scripts with file name pattern `tests/**/test_*_on_cpu.py` | |
| # - `gpu_unit_tests.yml`, run pytest on all scripts with file without the `on_cpu.py` suffix. | |
| # - Since cpu/gpu unit tests by default runs all tests under `tests`, please make sure tests are manually excluded in them when | |
| # - new workflow yaml is added to `.github/workflows` | |
| # - new tests are added to workflow mentioned in 2. | |
| name: e2e_ppo_trainer_megatron_vllm_2_ascend | |
| on: | |
| # Trigger the workflow on push or pull request, | |
| # but only for the main branch. | |
| # For push, for now only anti-patterns are specified so it is more conservative | |
| # and achieves higher coverage. | |
| push: | |
| branches: | |
| - main | |
| - v0.* | |
| paths: | |
| - "**/*.py" | |
| # Other entrypoints | |
| - "!verl/trainer/fsdp_sft_trainer.py" | |
| # FSDP | |
| - "!verl/workers/**/*dp_*.py" | |
| - "!verl/utils/fsdp_utils.py" | |
| - "!verl/utils/checkpoint/fsdp_checkpoint_manager.py" | |
| - "!verl/model_merger/fsdp_model_merger.py" | |
| pull_request: | |
| branches: | |
| - main | |
| - v0.* | |
| paths: | |
| - "**/*.py" | |
| # Other entrypoints | |
| - "!docker/**" | |
| # Docs | |
| - "!**/*.md" | |
| - "!docs/**" | |
| - "!examples/**" | |
| - "!tests/**" | |
| - "!verl/trainer/main_*.py" | |
| - "!verl/trainer/fsdp_sft_trainer.py" | |
| # FSDP | |
| - "!verl/workers/**/*dp_*.py" | |
| - "!verl/utils/fsdp_utils.py" | |
| - "!verl/utils/checkpoint/fsdp_checkpoint_manager.py" | |
| - "!verl/model_merger/fsdp_model_merger.py" | |
| # Entrypoints | |
| - ".github/workflows/e2e_ppo_trainer_megatron_vllm_2_ascend.yml" | |
| - "examples/data_preprocess/gsm8k.py" | |
| - "examples/data_preprocess/geo3k.py" | |
| - "tests/special_e2e/run_ppo_trainer_megatron.sh" | |
| - "verl/trainer/main_ppo.py" | |
| - "verl/trainer/config/ppo_megatron_trainer.yaml" | |
| # Cancel jobs on the same ref if a new one is triggered | |
| concurrency: | |
| group: ${{ github.workflow }}-${{ github.ref }} | |
| cancel-in-progress: ${{ github.ref != 'refs/heads/main' }} | |
| # Declare permissions just read content. | |
| permissions: | |
| contents: read | |
| jobs: | |
| e2e_ppo_trainer_fsdp_vllm_ascend: | |
| if: github.repository_owner == 'verl-project' | |
| runs-on: linux-aarch64-a2-8 | |
| timeout-minutes: 90 # Increase this timeout value as needed | |
| container: | |
| image: swr.ap-southeast-1.myhuaweicloud.com/base_image/ascend-ci/verl/verl:verl-8.3.rc1-910b-ubuntu22.04-py3.11-latest | |
| options: >- | |
| --shm-size 16g | |
| env: | |
| HF_ENDPOINT: "https://hf-mirror.com" | |
| HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable | |
| steps: | |
| - name: Check npu and CANN info | |
| run: | | |
| cat /usr/local/Ascend/ascend-toolkit/latest/"$(uname -i)"-linux/ascend_toolkit_install.info | |
| npu-smi info | |
| - name: Check initial pip list from image | |
| run: | | |
| pip list | |
| - name: Checkout verl-project/verl repo | |
| uses: actions/checkout@v4 | |
| with: | |
| fetch-depth: 0 | |
| clean: true | |
| - name: Install the current repository | |
| run: | | |
| pip install -r requirements-npu.txt | |
| pip install --no-deps -e . | |
| - name: Check final pip list | |
| run: | | |
| pip list | |
| - name: Prepare weights | |
| run: | | |
| ln -s /root/.cache/models ~/models | |
| - name: Prepare GSM8K dataset | |
| run: | | |
| python examples/data_preprocess/gsm8k.py --local_dataset_path ${HOME}/.cache/datasets/openai/gsm8k | |
| # Function RM | |
| - name: Running GSM8K E2E training tests on 8 L20 GPUs with rmpad using function rm with validation and saving (DDP_SIZE=2, FSDP_SIZE=4) | |
| run: | | |
| ray stop --force | |
| VAL_BEFORE_TRAIN=True TEST_FREQ=1 SAVE_FREQ=1 SAVE_HF_MODEL=True FSDP_SIZE=4 USE_KL=True VERL_EXP_NAME="qwen2.5-0.5b-function-reward-minimal-ddp-size2-fsdp-size4" bash tests/special_e2e/ppo_trainer/run_function_reward.sh | |
| - name: Test merging DDP+FSDP checkpoints (Qwen Actor) | |
| run: | | |
| exp_name="qwen2.5-0.5b-function-reward-minimal-ddp-size2-fsdp-size4" | |
| python -m verl.model_merger test --backend fsdp --local_dir checkpoints/verl-test/${exp_name}/global_step_1/actor --test_hf_dir checkpoints/verl-test/${exp_name}/global_step_1/actor/huggingface | |
| - name: Running GSM8K E2E training tests on 8 L20 GPUs with rmpad using function rm with validation and saving (FSDP2) | |
| run: | | |
| ray stop --force | |
| VAL_BEFORE_TRAIN=True TEST_FREQ=1 SAVE_FREQ=1 SAVE_HF_MODEL=True VERL_EXP_NAME="qwen2.5-0.5b-function-reward-minimal-fsdp2-size8" STRATEGY=fsdp2 bash tests/special_e2e/ppo_trainer/run_function_reward.sh | |
| - name: Test merging FSDP2 checkpoints (Qwen Actor) | |
| run: | | |
| exp_name="qwen2.5-0.5b-function-reward-minimal-fsdp2-size8" | |
| python -m verl.model_merger test --backend fsdp --local_dir checkpoints/verl-test/${exp_name}/global_step_1/actor --test_hf_dir checkpoints/verl-test/${exp_name}/global_step_1/actor/huggingface | |
| - name: Running GSM8K E2E without rmpad using function rm | |
| run: | | |
| ray stop --force | |
| RM_PAD=False bash tests/special_e2e/ppo_trainer/run_function_reward.sh | |
| - name: Running GSM8K E2E training tests on 8 L20 GPUs with rmpad using function rm (GRPO) | |
| run: | | |
| ray stop --force | |
| CUSTOM_REWARD_FN=True ADV_ESTIMATOR=grpo USE_KL=True bash tests/special_e2e/ppo_trainer/run_function_reward.sh | |
| # TODO | |
| # vllm0.11.0 not support them, will be opened until vllm0.13.0 ok | |
| # - name: Running GSM8K E2E training tests on 8 L20 GPUs with grpo lora using function rm with use_shm and layered_summon | |
| # run: | | |
| # ray stop --force | |
| # ADV_ESTIMATOR=grpo USE_SHM=True LORA_RANK=32 LOAD_FORMAT=safetensors LAYERED_SUMMON=True TOTAL_TRAIN_STEPS=1 SAVE_FREQ=1 FSDP_SIZE=4 VERL_EXP_NAME="qwen2.5-0.5b-function-reward-minimal" bash tests/special_e2e/ppo_trainer/run_function_reward.sh | |
| # - name: Test GRPO LoRA checkpoints merging function | |
| # run: | | |
| # export EXP_NAME="qwen2.5-0.5b-function-reward-minimal" | |
| # ls checkpoints/verl-test/${EXP_NAME}/global_step_1/actor | |
| # cat checkpoints/verl-test/${EXP_NAME}/global_step_1/actor/huggingface/config.json | |
| # python3 -m verl.model_merger merge --backend fsdp --local_dir checkpoints/verl-test/${EXP_NAME}/global_step_1/actor/ --target_dir checkpoints/verl-test/${EXP_NAME}/global_step_1/actor/huggingface | |
| # - name: Running GSM8K E2E training tests on 8 L20 GPUs with grpo lora using function rm with use_shm and layered_summon with fsdp2 | |
| # run: | | |
| # ray stop --force | |
| # ADV_ESTIMATOR=grpo USE_SHM=True LORA_RANK=32 LOAD_FORMAT=safetensors LAYERED_SUMMON=True STRATEGY=fsdp2 bash tests/special_e2e/ppo_trainer/run_function_reward.sh | |
| # TODO | |
| # vllm0.11 not support them, will be open after vllm0.13 | |
| # e2e_ppo_trainer_fsdp-qwen2_5vl-3b_ascend: | |
| # if: github.repository_owner == 'verl-project' | |
| # runs-on: linux-aarch64-a2-8 | |
| # timeout-minutes: 60 # Increase this timeout value as needed | |
| # container: | |
| # image: swr.ap-southeast-1.myhuaweicloud.com/base_image/ascend-ci/verl/verl:verl-8.3.rc1-910b-ubuntu22.04-py3.11-latest | |
| # options: >- | |
| # --shm-size 16g | |
| # env: | |
| # HF_ENDPOINT: "https://hf-mirror.com" | |
| # HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable | |
| # steps: | |
| # - name: Check npu and CANN info | |
| # run: | | |
| # cat /usr/local/Ascend/ascend-toolkit/latest/"$(uname -i)"-linux/ascend_toolkit_install.info | |
| # npu-smi info | |
| # - name: Check initial pip list from image | |
| # run: | | |
| # pip list | |
| # - name: Checkout verl-project/verl repo | |
| # uses: actions/checkout@v4 | |
| # with: | |
| # fetch-depth: 0 | |
| # clean: true | |
| # - name: Install the current repository | |
| # run: | | |
| # pip install -r requirements-npu.txt | |
| # pip install --no-deps -e . | |
| # pip install trl | |
| # - name: Check final pip list | |
| # run: | | |
| # pip list | |
| # - name: Prepare weights | |
| # run: | | |
| # ln -s /root/.cache/models ~/models | |
| # # Geo3k | |
| # - name: Prepare GEO3K dataset | |
| # run: | | |
| # python examples/data_preprocess/geo3k.py --local_dataset_path ${HOME}/.cache/datasets/hiyouga/geometry3k | |
| # - name: Running GEO3K VLM GRPO E2E training tests on 8 L20 GPUs with rmpad using function rm | |
| # run: | | |
| # ray stop --force | |
| # TRAIN_FILES=$HOME/data/geo3k/train.parquet VAL_FILES=$HOME/data/geo3k/test.parquet \ | |
| # MAX_PROMPT_LEN=1536 MAX_RESPONSE_LEN=1536 \ | |
| # MODEL_ID=Qwen/Qwen2.5-VL-3B-Instruct \ | |
| # ADV_ESTIMATOR=grpo RM_PAD=True USE_KL=True ENABLE_CHUNKED_PREFILL=False \ | |
| # SP_SIZE=2 \ | |
| # bash tests/special_e2e/ppo_trainer/run_function_reward.sh | |
| # - name: Running GEO3K VLM PPO E2E training tests on 8 L20 GPUs with rmpad using function rm | |
| # run: | | |
| # ray stop --force | |
| # TRAIN_FILES=$HOME/data/geo3k/train.parquet VAL_FILES=$HOME/data/geo3k/test.parquet \ | |
| # MAX_PROMPT_LEN=1536 MAX_RESPONSE_LEN=1536 \ | |
| # MODEL_ID=Qwen/Qwen2.5-VL-3B-Instruct \ | |
| # ADV_ESTIMATOR=gae RM_PAD=True USE_KL=True ENABLE_CHUNKED_PREFILL=False \ | |
| # SP_SIZE=2 \ | |
| # bash tests/special_e2e/ppo_trainer/run_function_reward.sh | |
| # - name: Running GEO3K VLM GRPO E2E lora training tests on 8 L20 GPUs with rmpad using function rm | |
| # run: | | |
| # ray stop --force | |
| # TRAIN_FILES=$HOME/data/geo3k/train.parquet VAL_FILES=$HOME/data/geo3k/test.parquet \ | |
| # MAX_PROMPT_LEN=1536 MAX_RESPONSE_LEN=1536 \ | |
| # MODEL_ID=Qwen/Qwen2.5-VL-3B-Instruct \ | |
| # ADV_ESTIMATOR=grpo RM_PAD=True USE_KL=True ENABLE_CHUNKED_PREFILL=False \ | |
| # SP_SIZE=2 \ | |
| # LORA_RANK=32 LORA_EXCLUDE=".*visual.*" \ | |
| # bash tests/special_e2e/ppo_trainer/run_function_reward.sh |