Files
NexQuant/rdagent/core/experiment.py
T

500 lines
19 KiB
Python
Raw Normal View History

2024-07-17 15:00:13 +08:00
from __future__ import annotations
import io
import os
2024-10-10 11:58:15 +08:00
import platform
import re
2024-07-17 15:00:13 +08:00
import shutil
import typing
2024-07-17 15:00:13 +08:00
import uuid
import zipfile
2024-06-14 12:59:44 +08:00
from abc import ABC, abstractmethod
from collections.abc import Sequence
2024-07-17 15:00:13 +08:00
from copy import deepcopy
from dataclasses import dataclass
from pathlib import Path
from typing import TYPE_CHECKING, Any, Generic, TypeVar
2024-07-17 15:00:13 +08:00
from rdagent.core.conf import RD_AGENT_SETTINGS
from rdagent.core.evaluation import Feedback
if TYPE_CHECKING:
from rdagent.utils.env import EnvResult
if typing.TYPE_CHECKING:
from rdagent.core.proposal import Hypothesis
from rdagent.utils.env import Env
2024-07-17 15:00:13 +08:00
"""
This file contains the all the class about organizing the task in RD-Agent.
"""
class AbsTask(ABC):
2024-09-23 19:53:05 +08:00
def __init__(self, name: str, version: int = 1) -> None:
"""
The version of the task, default is 1
Because qlib tasks execution and kaggle tasks execution are different, we need to distinguish them.
TODO: We may align them in the future.
"""
self.version = version
2024-09-23 19:53:05 +08:00
self.name = name
@abstractmethod
2024-07-18 22:36:04 +08:00
def get_task_information(self) -> str:
"""
Get the task information string to build the unique key
"""
2024-06-14 12:59:44 +08:00
class UserInstructions(list[str]):
def __str__(self) -> str:
if self:
return ("\nUser Instructions (Top priority!):\n" + "\n".join(f"- {ui}" for ui in self)) if self else ""
return ""
class Task(AbsTask):
def __init__(
self,
name: str,
version: int = 1,
description: str = "",
user_instructions: UserInstructions | None = None,
) -> None:
super().__init__(name, version)
self.description = description
self.user_instructions = user_instructions
def get_task_information(self) -> str:
return f"Task Name: {self.name}\nDescription: {self.description}{self.user_instructions!s}"
def __repr__(self) -> str:
return f"<{self.__class__.__name__} {self.name}>"
ASpecificTask = TypeVar("ASpecificTask", bound=Task)
ASpecificFeedback = TypeVar("ASpecificFeedback", bound=Feedback)
@dataclass
class RunningInfo:
result: object = None # The result of the experiment, can be different types in different scenarios.
running_time: float | None = None
class Workspace(ABC, Generic[ASpecificTask, ASpecificFeedback]):
2024-07-17 15:00:13 +08:00
"""
A workspace is a place to store the task implementation. It evolves as the developer implements the task.
To get a snapshot of the workspace, make sure call `copy` to get a copy of the workspace.
"""
2024-07-25 15:20:04 +08:00
def __init__(self, target_task: ASpecificTask | None = None) -> None:
self.target_task: ASpecificTask | None = target_task
self.feedback: ASpecificFeedback | None = None
self.running_info: RunningInfo = RunningInfo()
2024-06-14 12:59:44 +08:00
@abstractmethod
2024-07-25 15:20:04 +08:00
def execute(self, *args: Any, **kwargs: Any) -> object | None:
2024-07-18 22:36:04 +08:00
error_message = "execute method is not implemented."
raise NotImplementedError(error_message)
2024-07-17 15:00:13 +08:00
@abstractmethod
def copy(self) -> Workspace:
2024-07-18 22:36:04 +08:00
error_message = "copy method is not implemented."
raise NotImplementedError(error_message)
@property
@abstractmethod
def all_codes(self) -> str:
"""
Get all the code files in the workspace as a single string.
"""
# when the workspace is mutable inplace, provide support for creating checkpoints and recovering.
@abstractmethod
def create_ws_ckp(self) -> None:
"""
Create an in-memory checkpoint of the workspace so it can be restored later.
"""
@abstractmethod
def recover_ws_ckp(self) -> None:
"""
Restore the workspace from the checkpoint created by :py:meth:`create_ws_ckp`.
"""
2024-07-17 15:00:13 +08:00
ASpecificWS = TypeVar("ASpecificWS", bound=Workspace)
2024-07-17 15:00:13 +08:00
class WsLoader(ABC, Generic[ASpecificTask, ASpecificWS]):
@abstractmethod
2024-07-17 15:00:13 +08:00
def load(self, task: ASpecificTask) -> ASpecificWS:
2024-07-18 22:36:04 +08:00
error_message = "load method is not implemented."
raise NotImplementedError(error_message)
2024-07-17 15:00:13 +08:00
class FBWorkspace(Workspace):
"""
2024-07-17 15:00:13 +08:00
File-based task workspace
The implemented task will be a folder which contains related elements.
- Data
2024-07-17 15:00:13 +08:00
- Code Workspace
- Output
- After execution, it will generate the final output as file.
2024-07-18 22:36:04 +08:00
A typical way to run the pipeline of FBWorkspace will be:
(We didn't add it as a method due to that we may pass arguments into
`prepare` or `execute` based on our requirements.)
.. code-block:: python
def run_pipeline(self, **files: str):
self.prepare()
self.inject_files(**files)
self.execute()
"""
2024-07-18 22:36:04 +08:00
def __init__(self, *args: Any, **kwargs: Any) -> None:
super().__init__(*args, **kwargs)
self.file_dict: dict[str, Any] = (
2024-07-17 15:00:13 +08:00
{}
) # The code injected into the folder, store them in the variable to reproduce the former result
self.workspace_path: Path = RD_AGENT_SETTINGS.workspace_path / uuid.uuid4().hex
self.ws_ckp: bytes | None = None # In-memory checkpoint data created by ``create_ws_ckp``.
self.change_summary: str | None = None # The change from the previous version of workspace
@staticmethod
def _format_code_dict(code_dict: dict[str, str]) -> str:
"""
Helper function to format the code dictionary into a string.
"""
code_string = ""
2025-02-14 16:24:26 +08:00
for file_name in sorted(code_dict.keys()):
code_string += f"\nFile Path: {file_name}\n```\n{code_dict[file_name]}\n```"
return code_string
@property
def all_codes(self) -> str:
"""
Get all the code files in the workspace as a single string, excluding test files.
"""
filtered_dict = {k: v for k, v in self.file_dict.items() if k.endswith(".py") and "test" not in k}
return self._format_code_dict(filtered_dict)
def get_codes(self, pattern: str) -> str:
"""
Get code files matching a specific pattern as a single string, excluding test files.
"""
filtered_dict = {
k: v for k, v in self.file_dict.items() if re.search(pattern, k) and k.endswith(".py") and "test" not in k
}
return self._format_code_dict(filtered_dict)
2024-07-18 22:36:04 +08:00
def prepare(self) -> None:
"""
2024-07-17 15:00:13 +08:00
Prepare the workspace except the injected code
- Data
- Documentation
typical usage of `*args, **kwargs`:
Different methods shares the same data. The data are passed by the arguments.
"""
2024-07-17 15:00:13 +08:00
self.workspace_path.mkdir(parents=True, exist_ok=True)
@staticmethod
def link_all_files_in_folder_to_workspace(data_path: Path, workspace_path: Path) -> None:
data_path = Path(data_path).absolute() # in case of relative path that will be invalid when we change cwd.
workspace_path = Path(workspace_path)
for data_file_path in data_path.iterdir():
workspace_data_file_path = workspace_path / data_file_path.name
if workspace_data_file_path.exists():
workspace_data_file_path.unlink()
if platform.system() in ("Linux", "Darwin"):
2025-09-11 16:14:48 +08:00
workspace_data_file_path.symlink_to(data_file_path)
2024-10-10 11:58:15 +08:00
if platform.system() == "Windows":
os.link(data_file_path, workspace_data_file_path)
DEL_KEY = "__DEL__"
def inject_files(self, **files: str) -> None:
"""
Inject the code into the folder.
{
<file name1>: <code>, // indicate writing <code> into <file name>
(create new file or replace existing file)
<file name2>: "__DEL__" // indicate removing file name2. When we want to replace a file to a new one,
we usually use this
}
"""
2024-07-17 15:00:13 +08:00
self.prepare()
for k, v in files.items():
target_file_path = self.workspace_path / k # Define target_file_path before using it
if v == self.DEL_KEY: # Use self.DEL_KEY to access the class variable
if target_file_path.exists():
target_file_path.unlink() # Unlink the file if it exists
self.file_dict.pop(k, None) # Safely remove the key from file_dict
else:
self.file_dict[k] = v
target_file_path.parent.mkdir(parents=True, exist_ok=True)
target_file_path.write_text(v)
2026-03-02 19:04:10 +08:00
def remove_files(self, file_names: str | list[str]) -> None:
"""
Remove specified files from the workspace.
"""
if isinstance(file_names, str):
file_names = [file_names]
for file_name in file_names:
target_file_path = self.workspace_path / file_name
if target_file_path.exists():
target_file_path.unlink() # Unlink the file if it exists
self.file_dict.pop(file_name, None) # Safely remove the key from file_dict
def get_files(self) -> list[Path]:
"""
Get the environment description.
To be general, we only return a list of filenames.
2024-07-17 15:00:13 +08:00
How to summarize the environment is the responsibility of the Developer.
"""
return list(self.workspace_path.iterdir())
2024-06-14 12:59:44 +08:00
2024-07-18 22:36:04 +08:00
def inject_code_from_folder(self, folder_path: Path) -> None:
2024-07-17 15:00:13 +08:00
"""
Load the workspace from the folder
"""
for file_path in folder_path.rglob("*"):
if file_path.suffix in (".py", ".yaml", ".md"):
relative_path = file_path.relative_to(folder_path)
self.inject_files(**{str(relative_path): file_path.read_text()})
2024-07-17 15:00:13 +08:00
def inject_code_from_file_dict(self, workspace: FBWorkspace) -> None:
"""
Load the workspace from the file_dict
"""
2026-03-02 19:04:10 +08:00
# NOTE: this is a deprecated method, use inject_from_workspace instead
# TODO: remove this method; it is only for compatibility with old codes
self.inject_from_workspace(workspace)
def inject_from_workspace(self, workspace: FBWorkspace) -> None:
for name, code in workspace.file_dict.items():
self.inject_files(**{name: code})
2024-07-17 15:00:13 +08:00
def copy(self) -> FBWorkspace:
"""
copy the workspace from the original one
"""
return deepcopy(self)
def clear(self) -> None:
"""
Clear the workspace
"""
shutil.rmtree(self.workspace_path, ignore_errors=True)
self.file_dict = {}
2024-07-17 15:00:13 +08:00
2025-02-25 12:06:07 +08:00
def before_execute(self) -> None:
"""
Before executing the code, we need to prepare the workspace and inject code into the workspace.
"""
self.prepare()
self.inject_files(**self.file_dict)
def execute(self, env: Env, entry: str) -> str:
2024-07-17 15:00:13 +08:00
"""
Before each execution, make sure to prepare and inject code.
"""
result = self.run(env, entry)
2026-03-02 19:04:10 +08:00
return result.stdout # NOTE: truncating just for aligning with the old code.
def run(self, env: Env, entry: str) -> EnvResult:
"""
Execute the code in the environment and return an EnvResult object (stdout, exit_code, running_time).
Before each execution, make sure to prepare and inject code.
2024-07-17 15:00:13 +08:00
"""
self.prepare()
self.inject_files(**self.file_dict)
return env.run(entry, str(self.workspace_path), env={"PYTHONPATH": "./"})
2024-07-17 15:00:13 +08:00
def create_ws_ckp(self) -> None:
"""
Zip the contents of ``workspace_path`` and persist the archive on
``self.ws_ckp`` for later restoration via :py:meth:`recover_ws_ckp`.
"""
buf = io.BytesIO()
with zipfile.ZipFile(buf, "w", zipfile.ZIP_DEFLATED) as zf:
for file_path in self.workspace_path.rglob("*"):
# Only include regular files up to 100 KB so that the checkpoint
# remains lightweight. Larger files (for example, datasets) are
# expected to be recreated or mounted separately.
if file_path.is_symlink():
# Preserve symbolic links within the archive
zi = zipfile.ZipInfo(str(file_path.relative_to(self.workspace_path)))
zi.create_system = 3 # indicates Unix
zi.external_attr = 0o120777 << 16 # symlink file type + 0777 perms
zf.writestr(zi, str(file_path.readlink()))
elif file_path.is_file():
size_limit = RD_AGENT_SETTINGS.workspace_ckp_size_limit
if (
RD_AGENT_SETTINGS.workspace_ckp_white_list_names is not None
and file_path.name in RD_AGENT_SETTINGS.workspace_ckp_white_list_names
) or (size_limit <= 0 or file_path.stat().st_size <= size_limit):
zf.write(file_path, file_path.relative_to(self.workspace_path))
self.ws_ckp = buf.getvalue()
def recover_ws_ckp(self) -> None:
"""
Restore the workspace directory from the in-memory checkpoint created by
:py:meth:`create_ws_ckp`.
"""
if self.ws_ckp is None:
msg = "Workspace checkpoint doesn't exist. Call `create_ws_ckp` first."
raise RuntimeError(msg)
shutil.rmtree(self.workspace_path, ignore_errors=True)
self.workspace_path.mkdir(parents=True, exist_ok=True)
buf = io.BytesIO(self.ws_ckp)
with zipfile.ZipFile(buf, "r") as zf:
for info in zf.infolist():
dest_path = self.workspace_path / info.filename
# File type bits (upper 4) are in high 16 bits of external_attr
mode = (info.external_attr >> 16) & 0o170000
symlink_mode = 0o120000 # Constant for symlink file type in Unix
if mode == symlink_mode: # Symlink
dest_path.parent.mkdir(parents=True, exist_ok=True)
link_target = zf.read(info).decode()
2025-09-11 16:14:48 +08:00
dest_path.symlink_to(link_target)
elif info.is_dir():
dest_path.mkdir(parents=True, exist_ok=True)
else:
dest_path.parent.mkdir(parents=True, exist_ok=True)
with dest_path.open("wb") as f:
f.write(zf.read(info))
# NOTE: very important to reduce the size of the object
self.ws_ckp = None
2024-09-29 09:59:54 +08:00
def __str__(self) -> str:
return f"Workspace[{self.workspace_path=}" + (
"]" if self.target_task is None else f",{self.target_task.name=}]"
)
2024-07-17 15:00:13 +08:00
ASpecificWSForExperiment = TypeVar("ASpecificWSForExperiment", bound=Workspace)
ASpecificWSForSubTasks = TypeVar("ASpecificWSForSubTasks", bound=Workspace)
2025-07-23 12:20:06 +08:00
class ExperimentPlan(dict[str, Any]):
"""
A plan for the experiment, which is a dictionary that contains the plan to each stage.
"""
class Experiment(
ABC,
Generic[ASpecificTask, ASpecificWSForExperiment, ASpecificWSForSubTasks],
):
"""
2024-07-17 15:00:13 +08:00
The experiment is a sequence of tasks and the implementations of the tasks after generated by the Developer.
"""
def __init__(
self,
sub_tasks: Sequence[ASpecificTask],
based_experiments: Sequence[ASpecificWSForExperiment] = [],
hypothesis: Hypothesis | None = None,
) -> None:
self.hypothesis: Hypothesis | None = hypothesis # Experiment is optionally generated by hypothesis
self.sub_tasks: Sequence[ASpecificTask] = sub_tasks
# None means
# - initialization placeholder before implementation
# - the developer actively skip the task;
2024-07-25 15:20:04 +08:00
self.sub_workspace_list: list[ASpecificWSForSubTasks | None] = [None] * len(self.sub_tasks)
# TODO:
# It will be used in runner in history
# If we implement the whole workflow, we don't have to use it, then we remove it.
self.based_experiments: Sequence[ASpecificWSForExperiment] = based_experiments
self.experiment_workspace: ASpecificWSForExperiment | None = None
# The experiment may be developed by different developers.
# Last feedback is used to propagate info to the next developer.
# Life cycle:
# - Developer assigns feedback for next component;
# - Workflow control clears feedback.
self.prop_dev_feedback: Feedback | None = None
# TODO: (xiao) I think this is too concrete; we should move it into
# NOTE: Assumption
# - only runner will assign this variable
# - We will always create a new Experiment without copying previous results when we goto the next new loop.
self.running_info = RunningInfo()
self.sub_results: dict[str, float] = (
{}
) # TODO: in Kaggle, now sub results are all saved in self.result, remove this in the future.
# For parallel multi-trace support
self.local_selection: tuple[int, ...] | None = None
2025-07-23 12:20:06 +08:00
self.plan: ExperimentPlan | None = (
None # To store the planning information for this experiment, should be generated inside exp_gen.gen
)
self.user_instructions: UserInstructions | None = None # To store the user instructions for this experiment
def set_user_instructions(self, user_instructions: UserInstructions | None) -> None:
if user_instructions is None:
return
if not isinstance(user_instructions, UserInstructions) and isinstance(user_instructions, list):
user_instructions = UserInstructions(user_instructions)
self.user_instructions = user_instructions
for ws in self.sub_workspace_list:
if ws is not None:
ws.target_task.user_instructions = user_instructions # type: ignore[union-attr]
for task in self.sub_tasks:
task.user_instructions = user_instructions
if self.experiment_workspace is not None and self.experiment_workspace.target_task is not None:
self.experiment_workspace.target_task.user_instructions = user_instructions
@property
def result(self) -> object:
return self.running_info.result
@result.setter
def result(self, value: object) -> None:
self.running_info.result = value
# when the workspace is mutable inplace, provide support for creating checkpoints and recovering.
def create_ws_ckp(self) -> None:
if self.experiment_workspace is not None:
self.experiment_workspace.create_ws_ckp()
for ws in self.sub_workspace_list:
if ws is not None:
ws.create_ws_ckp()
def recover_ws_ckp(self) -> None:
if self.experiment_workspace is not None:
self.experiment_workspace.recover_ws_ckp()
for ws in self.sub_workspace_list:
if ws is not None:
try:
ws.recover_ws_ckp()
except RuntimeError:
# the FBWorkspace is shared between experiment_workspace and sub_workspace_list,
# so recover_ws_ckp will raise RuntimeError if a workspace is recovered twice.
print("recover_ws_ckp failed due to one workspace is recovered twice.")
ASpecificExp = TypeVar("ASpecificExp", bound=Experiment)
2025-07-23 12:20:06 +08:00
ASpecificPlan = TypeVar("ASpecificPlan", bound=ExperimentPlan)
TaskOrExperiment = TypeVar("TaskOrExperiment", Task, Experiment)
2024-06-14 12:59:44 +08:00
class Loader(ABC, Generic[TaskOrExperiment]):
@abstractmethod
2024-07-18 22:36:04 +08:00
def load(self, *args: Any, **kwargs: Any) -> TaskOrExperiment:
err_msg = "load method is not implemented."
raise NotImplementedError(err_msg)