Skip to content

Could not find 'data/legacy_data' #7

Description

@xingzhejun

Hi,

Thank you for your work!

I met one issue before starting training.

Traceback (most recent call last):
File "", line 198, in _run_module_as_main
File "", line 88, in _run_code
File "/users/yifuluo/projects/RLCSD-main/src/self_distill_main.py", line 1098, in
main()
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/main.py", line 94, in decorated_main
_run_hydra(
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/utils.py", line 394, in _run_hydra
_run_app(
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/utils.py", line 457, in _run_app
run_and_report(
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/utils.py", line 223, in run_and_report
raise ex
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/utils.py", line 220, in run_and_report
return func()
^^^^^^
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/utils.py", line 458, in
lambda: hydra.run(
^^^^^^^^^^
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/hydra.py", line 105, in run
cfg = self.compose_config(
^^^^^^^^^^^^^^^^^^^^
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/hydra.py", line 594, in compose_config
cfg = self.config_loader.load_configuration(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/config_loader_impl.py", line 142, in load_configuration
return self._load_configuration_impl(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/config_loader_impl.py", line 253, in _load_configuration_impl
defaults_list = create_defaults_list(
^^^^^^^^^^^^^^^^^^^^^
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/defaults_list.py", line 745, in create_defaults_list
defaults, tree = _create_defaults_list(
^^^^^^^^^^^^^^^^^^^^^^
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/defaults_list.py", line 715, in _create_defaults_list
defaults_tree = _create_defaults_tree(
^^^^^^^^^^^^^^^^^^^^^^
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/defaults_list.py", line 356, in _create_defaults_tree
ret = _create_defaults_tree_impl(
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/defaults_list.py", line 457, in _create_defaults_tree_impl
return _expand_virtual_root(repo, root, overrides, skip_missing)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/defaults_list.py", line 280, in _expand_virtual_root
subtree = _create_defaults_tree_impl(
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/defaults_list.py", line 573, in _create_defaults_tree_impl
add_child(children, new_root)
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/internal/defaults_list.py", line 520, in add_child
subtree
= _create_defaults_tree_impl(
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/defaults_list.py", line 488, in _create_defaults_tree_impl
config_not_found_error(repo=repo, tree=root)
File "/tmp/yifuluo/python-envs/dev/lib/python3.12/site-packages/hydra/_internal/defaults_list.py", line 799, in config_not_found_error
raise MissingConfigException(
hydra.errors.MissingConfigException: In 'ppo_trainer': Could not find 'data/legacy_data'

I asked codex, and it said:

self_distill_main.py 使用:
@hydra.main(
config_path="../third_party/verl/verl/trainer/config",
config_name="ppo_trainer",
version_base=None,
)
因此 Hydra 首先加载:
ppo_trainer.yaml (line 8)
该文件的 defaults 中写了:
defaults:

  • model_engine: dp
  • actor@actor_rollout_ref.actor: ${model_engine}_actor
  • data@data: legacy_data
    其中:
  • data@data: legacy_data
    是 Hydra 的 config group 语法,含义是:
    读取:
    third_party/verl/verl/trainer/config/data/legacy_data.yaml

然后把内容挂载到最终配置的 data 字段
但是当前仓库里:
third_party/verl/verl/trainer/config/data/
整个目录都不存在,因此自然也没有:
third_party/verl/verl/trainer/config/data/legacy_data.yaml
所以 Hydra 报:
Could not find 'data/legacy_data'

I checked there is indeed no this folder. Any suggestions? Thank you!

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions