Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

use toml's in file config store #100

Merged
merged 1 commit into from
Feb 14, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
57 changes: 57 additions & 0 deletions dacapo/store/config_store.py
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,45 @@ class DuplicateNameError(Exception):
class ConfigStore(ABC):
"""Base class for configuration stores."""

@property
@abstractmethod
def runs(self):
pass

@property
@abstractmethod
def datasplits(self):
pass

@property
@abstractmethod
def datasets(self):
pass

@property
@abstractmethod
def arrays(self):
pass

@property
@abstractmethod
def tasks(self):
pass

@property
@abstractmethod
def trainers(self):
pass

@property
@abstractmethod
def architectures(self):
pass

@abstractmethod
def delete_config(self, database, config_name: str) -> None:
pass

@abstractmethod
def store_run_config(self, run_config: "RunConfig") -> None:
"""Store a run config. This should also store the configs that are part
Expand All @@ -34,6 +73,9 @@ def retrieve_run_config_names(self) -> List[str]:
"""Retrieve all run config names."""
pass

def delete_run_config(self, run_name: str) -> None:
self.delete_config(self.runs, run_name)

@abstractmethod
def store_task_config(self, task_config: "TaskConfig") -> None:
"""Store a task config."""
Expand All @@ -49,6 +91,9 @@ def retrieve_task_config_names(self) -> List[str]:
"""Retrieve all task config names."""
pass

def delete_task_config(self, task_name: str) -> None:
self.delete_config(self.tasks, task_name)

@abstractmethod
def store_architecture_config(
self, architecture_config: "ArchitectureConfig"
Expand All @@ -68,6 +113,9 @@ def retrieve_architecture_config_names(self) -> List[str]:
"""Retrieve all architecture config names."""
pass

def delete_architecture_config(self, architecture_name: str) -> None:
self.delete_config(self.architectures, architecture_name)

@abstractmethod
def store_trainer_config(self, trainer_config: "TrainerConfig") -> None:
"""Store a trainer config."""
Expand All @@ -83,6 +131,9 @@ def retrieve_trainer_config_names(self) -> List[str]:
"""Retrieve all trainer config names."""
pass

def delete_trainer_config(self, trainer_name: str) -> None:
self.delete_config(self.trainers, trainer_name)

@abstractmethod
def store_datasplit_config(self, datasplit_config: "DataSplitConfig") -> None:
"""Store a datasplit config."""
Expand All @@ -98,6 +149,9 @@ def retrieve_datasplit_config_names(self) -> List[str]:
"""Retrieve all datasplit names."""
pass

def delete_datasplit_config(self, datasplit_name: str) -> None:
self.delete_config(self.datasplits, datasplit_name)

@abstractmethod
def store_array_config(self, array_config: "ArrayConfig") -> None:
"""Store a array config."""
Expand All @@ -112,3 +166,6 @@ def retrieve_array_config(self, array_name: str) -> "ArrayConfig":
def retrieve_array_config_names(self) -> List[str]:
"""Retrieve all array names."""
pass

def delete_array_config(self, array_name: str) -> None:
self.delete_config(self.arrays, array_name)
69 changes: 47 additions & 22 deletions dacapo/store/file_config_store.py
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,7 @@
from dacapo.experiments.trainers import TrainerConfig

import logging
import pickle
import toml
from pathlib import Path

logger = logging.getLogger(__name__)
Expand Down Expand Up @@ -36,7 +36,7 @@ def retrieve_run_config(self, run_name):
return converter.structure(run_doc, RunConfig)

def retrieve_run_config_names(self):
return [f.name for f in self.runs.iterdir()]
return [f.name[:-5] for f in self.runs.iterdir()]

def store_task_config(self, task_config):
task_doc = converter.unstructure(task_config)
Expand All @@ -47,7 +47,7 @@ def retrieve_task_config(self, task_name):
return converter.structure(task_doc, TaskConfig)

def retrieve_task_config_names(self):
return [f.name for f in self.tasks.iterdir()]
return [f.name[:-5] for f in self.tasks.iterdir()]

def store_architecture_config(self, architecture_config):
architecture_doc = converter.unstructure(architecture_config)
Expand All @@ -58,7 +58,7 @@ def retrieve_architecture_config(self, architecture_name):
return converter.structure(architecture_doc, ArchitectureConfig)

def retrieve_architecture_config_names(self):
return [f.name for f in self.architectures.iterdir()]
return [f.name[:-5] for f in self.architectures.iterdir()]

def store_trainer_config(self, trainer_config):
trainer_doc = converter.unstructure(trainer_config)
Expand All @@ -69,7 +69,7 @@ def retrieve_trainer_config(self, trainer_name):
return converter.structure(trainer_doc, TrainerConfig)

def retrieve_trainer_config_names(self):
return [f.name for f in self.trainers.iterdir()]
return [f.name[:-5] for f in self.trainers.iterdir()]

def store_datasplit_config(self, datasplit_config):
datasplit_doc = converter.unstructure(datasplit_config)
Expand All @@ -80,7 +80,7 @@ def retrieve_datasplit_config(self, datasplit_name):
return converter.structure(datasplit_doc, DataSplitConfig)

def retrieve_datasplit_config_names(self):
return [f.name for f in self.datasplits.iterdir()]
return [f.name[:-5] for f in self.datasplits.iterdir()]

def store_array_config(self, array_config):
array_doc = converter.unstructure(array_config)
Expand All @@ -91,19 +91,17 @@ def retrieve_array_config(self, array_name):
return converter.structure(array_doc, ArrayConfig)

def retrieve_array_config_names(self):
return [f.name for f in self.arrays.iterdir()]
return [f.name[:-5] for f in self.arrays.iterdir()]

def __save_insert(self, collection, data, ignore=None):
name = data["name"]

file_store = collection / name
file_store = collection / f"{name}.toml"
if not file_store.exists():
with file_store.open("wb") as fd:
pickle.dump(dict(data), fd)
toml.dump(dict(data), file_store.open("w"))

else:
with file_store.open("rb") as fd:
existing = pickle.load(fd)
existing = toml.load(file_store.open("r"))

if not self.__same_doc(existing, data, ignore):
raise DuplicateNameError(
Expand All @@ -113,10 +111,9 @@ def __save_insert(self, collection, data, ignore=None):
)

def __load(self, collection, name):
file_store = collection / name
file_store = collection / f"{name}.toml"
if file_store.exists():
with file_store.open("rb") as fd:
return pickle.load(fd)
return toml.load(file_store.open("r"))
else:
raise ValueError(f"No config with name: {name} in collection: {collection}")

Expand All @@ -138,17 +135,45 @@ def __init_db(self):
pass

def __open_collections(self):
self.users = self.path / "users"
self.users.mkdir(exist_ok=True, parents=True)
self.runs = self.path / "runs"
self.runs.mkdir(exist_ok=True, parents=True)
self.tasks = self.path / "tasks"
self.tasks.mkdir(exist_ok=True, parents=True)
self.datasplits = self.path / "datasplits"
self.datasplits.mkdir(exist_ok=True, parents=True)
self.arrays = self.path / "arrays"
self.arrays.mkdir(exist_ok=True, parents=True)
self.architectures = self.path / "architectures"
self.architectures.mkdir(exist_ok=True, parents=True)
self.trainers = self.path / "trainers"
self.trainers.mkdir(exist_ok=True, parents=True)

@property
def users(self) -> Path:
return self.path / "users"

@property
def runs(self) -> Path:
return self.path / "runs"

@property
def tasks(self) -> Path:
return self.path / "tasks"

@property
def datasplits(self) -> Path:
return self.path / "datasplits"

@property
def arrays(self) -> Path:
return self.path / "arrays"

@property
def architectures(self) -> Path:
return self.path / "architectures"

@property
def trainers(self) -> Path:
return self.path / "trainers"

@property
def datasets(self) -> Path:
return self.path / "datasets"

def delete_config(self, database: Path, config_name: str) -> None:
(database / f"{config_name}.toml").unlink()
Loading