feat!: remove deprecated classes and files

crunch/__init__.py

@@ -8,5 +8,4 @@
 import crunch.api as api
 import crunch.store as store
 from crunch.inline import load as load_notebook
-from crunch.container import Columns as Columns
 from crunch.runner import is_inside as is_inside_runner

crunch/api/__init__.py

@@ -6,21 +6,15 @@
 from crunch.api.domain.competition import CompetitionMode as CompetitionMode
 from crunch.api.domain.competition import CompetitionStatus as CompetitionStatus
 from crunch.api.domain.crunch import Crunch as Crunch
-from crunch.api.domain.data_release import ColumnNames as ColumnNames
 from crunch.api.domain.data_release import DataFile as DataFile
 from crunch.api.domain.data_release import DataFiles as DataFiles
-from crunch.api.domain.data_release import DataFilesUnion as DataFilesUnion
 from crunch.api.domain.data_release import DataRelease as DataRelease
-from crunch.api.domain.data_release import DataReleaseFeature as DataReleaseFeature
 from crunch.api.domain.data_release import DataReleaseSplit as DataReleaseSplit
 from crunch.api.domain.data_release import DataReleaseSplitGroup as DataReleaseSplitGroup
 from crunch.api.domain.data_release import DataReleaseSplitReduced as DataReleaseSplitReduced
 from crunch.api.domain.data_release import DataReleaseTargetResolution as DataReleaseTargetResolution
-from crunch.api.domain.data_release import KnownData as KnownData
-from crunch.api.domain.data_release import OriginalFiles as OriginalFiles
 from crunch.api.domain.data_release import SizeVariant as SizeVariant
 from crunch.api.domain.data_release import SplitKeyPythonType as SplitKeyPythonType
-from crunch.api.domain.data_release import TargetColumnNames as TargetColumnNames
 from crunch.api.domain.enum_ import Language as Language
 from crunch.api.domain.enum_ import SplitKeyType as SplitKeyType
 from crunch.api.domain.leaderboard import Leaderboard as Leaderboard

crunch/api/domain/data_release.py

@@ -1,7 +1,6 @@
 from dataclasses import dataclass
 from enum import Enum
-from types import MappingProxyType
-from typing import TYPE_CHECKING, Any, Dict, Iterator, List, Optional, Tuple, Union
+from typing import TYPE_CHECKING, Any, Dict, Iterator, List, Optional, Union
 
 from dataclasses_json import LetterCase, Undefined, dataclass_json
 
@@ -37,53 +36,7 @@ class DataFile:
     compressed: bool
 
 
-@dataclass(frozen=True)
-class KnownData:
-
-    X_TRAIN = "x_train"
-    Y_TRAIN = "y_train"
-    X_TEST = "x_test"
-    Y_TEST = "y_test"
-    X = "x"
-    Y = "y"
-    Y_RAW = "y_raw"
-    EXAMPLE_PREDICTION = "example_prediction"
-
-
-@dataclass_json(
-    letter_case=LetterCase.CAMEL,
-    undefined=Undefined.EXCLUDE
-)
-@dataclass(frozen=True)
-class DataFiles:
-
-    x_train: DataFile
-    x_test: DataFile
-    y_train: DataFile
-    y_test: DataFile
-    example_prediction: DataFile
-
-    def items(self):
-        return vars(self).items()
-
-
-@dataclass_json(
-    letter_case=LetterCase.CAMEL,
-    undefined=Undefined.EXCLUDE
-)
-@dataclass(frozen=True)
-class OriginalFiles:
-
-    x: DataFile
-    y: DataFile
-    y_raw: Optional[DataFile]
-    example_prediction: DataFile
-
-    def items(self):
-        return vars(self).items()
-
-
-DataFilesUnion = Union[DataFiles, OriginalFiles, Dict[str, DataFile]]
+DataFiles = Dict[str, DataFile]
 
 
 class DataReleaseSplitGroup(Enum):
@@ -128,26 +81,6 @@ def from_dict_array(
         ]
 
 
-@dataclass_json(
-    letter_case=LetterCase.CAMEL,
-    undefined=Undefined.EXCLUDE,
-)
-@dataclass(frozen=True)
-class DataReleaseFeature:
-
-    group: str
-    name: str
-
-    @staticmethod
-    def from_dict_array(
-        input: List[Dict[str, Any]]
-    ):
-        return [
-            DataReleaseFeature.from_dict(x)
-            for x in input
-        ]
-
-
 class SizeVariant(Enum):
 
     DEFAULT = "DEFAULT"
@@ -193,10 +126,6 @@ def number(self) -> int:
     def embargo(self) -> int:
         return self._attrs["embargo"]
 
-    @property
-    def column_names(self) -> "ColumnNames":
-        return ColumnNames.from_dict(self._attrs["columnNames"])
-
     @property
     def number_of_features(self) -> int:
         return self._attrs["numberOfFeatures"]
@@ -210,22 +139,16 @@ def target_resolution(self):
         return DataReleaseTargetResolution[self._attrs["target_resolution"]]
 
     @property
-    def data_files(self) -> DataFilesUnion:
+    def data_files(self) -> DataFiles:
         files = self._attrs.get("dataFiles")
         if not files:
             self.reload()
             files = self._attrs["dataFiles"]
 
-        if "xTrain" in files and "yTrain" in files:
-            return DataFiles.from_dict(files)
-
-        if "x" in files and "y" in files:
-            return OriginalFiles.from_dict(files)
-
-        return MappingProxyType({
+        return {
             key: DataFile.from_dict(value)
             for key, value in files.items()
-        })
+        }
 
     @property
     def splits(self) -> List[DataReleaseSplit]:
@@ -236,92 +159,15 @@ def splits(self) -> List[DataReleaseSplit]:
 
         return list(DataReleaseSplit.from_dict_array(splits))
 
-    @property
-    def default_feature_group(self) -> str:
-        return self._attrs["defaultFeatureGroup"]
-
-    @property
-    def features(self) -> Tuple[DataReleaseFeature]:
-        features = self._attrs.get("features")
-        if features is None:
-            self.reload(include_features=True)
-            features = self._attrs["features"]
-
-        return tuple(DataReleaseFeature.from_dict_array(features))
-
     def reload(
         self,
         include_splits: bool = True,
-        include_features: bool = True,
     ):
         return super().reload(
             include_splits=include_splits,
-            include_features=include_features,
         )
 
 
-@dataclass_json(
-    letter_case=LetterCase.CAMEL,
-    undefined=Undefined.EXCLUDE
-)
-@dataclass(frozen=True)
-class TargetColumnNames:
-
-    id: int
-    name: str
-    side: Optional[str]
-    input: Optional[str]
-    output: Optional[str]
-    file_path: Optional[str]
-
-
-@dataclass_json(
-    letter_case=LetterCase.CAMEL,
-    undefined=Undefined.EXCLUDE
-)
-@dataclass(frozen=True)
-class ColumnNames:
-
-    id: Optional[str]
-    moon: Optional[str]
-    side: Optional[str]
-    input: Optional[str]
-    output: Optional[str]
-    targets: List[TargetColumnNames]
-
-    @property
-    def first_target(self):
-        return next(iter(self.targets), None)
-
-    @property
-    def inputs(self):
-        return [
-            target_column_names.input
-            for target_column_names in self.targets
-        ]
-
-    @property
-    def outputs(self):
-        return [
-            target_column_names.output
-            for target_column_names in self.targets
-        ]
-
-    @property
-    def target_names(self):
-        return [
-            target_column_names.name
-            for target_column_names in self.targets
-        ]
-
-    def get_target_by_name(self, name: str):
-        for target in self.targets:
-            if target.name == name:
-                return target
-
-        return None
-
-
 class DataReleaseCollection(Collection[DataRelease]):
 
     model = DataRelease
@@ -342,14 +188,12 @@ def get(
         self,
         number: Union[int, str],
         include_splits: bool = True,
-        include_features: bool = True,
     ) -> DataRelease:
         return self.prepare_model(
             self._client.api.get_data_release(
                 self.competition.id,
                 number,
                 include_splits=include_splits,
-                include_features=include_features,
             )
         )
 
@@ -387,14 +231,12 @@ def get_data_release(
         competition_identifier: "CompetitionIdentifierType",
         number: int,
         include_splits: bool = False,
-        include_features: bool = False,
     ):
         return self._result(
             self.get(
                 f"/v1/competitions/{competition_identifier}/data-releases/{number}",
                 params={
                     "includeSplits": include_splits,
-                    "includeFeatures": include_features,
                 }
             ),
             json=True

crunch/api/errors.py

@@ -5,7 +5,8 @@
 from crunch.api.domain.phase import PhaseType
 from crunch.api.domain.project import ProjectTokenType
 from crunch.external.inflection import camelize, underscore
-from crunch.utils import format_bytes, smart_call, try_get_competition_name
+from crunch.external.humanfriendly import format_size
+from crunch.utils import smart_call, try_get_competition_name
 
 
 def _print_contact(
@@ -343,8 +344,8 @@ def print_helper(
         self,
         **kwargs: Any,
     ) -> None:
-        size = format_bytes(self.size) if self.size is not None else None
-        maximum_size = format_bytes(self.maximum_size)
+        size = format_size(self.size) if self.size is not None else None
+        maximum_size = format_size(self.maximum_size)
 
         sizes = f"{size}/{maximum_size}" if size is not None else f"max. {maximum_size}"
         print(f"The resources directory is too big. ({sizes})")
@@ -406,7 +407,7 @@ def print_helper(
         self,
         **kwargs: Any,
     ) -> None:
-        print(f"Prediction is too big. ({format_bytes(self.size)}/{format_bytes(self.maximum_size)})")
+        print(f"Prediction is too big. ({format_size(self.size)}/{format_size(self.maximum_size)})")
 
 
 class ProjectNotFoundException(ApiException):
@@ -578,7 +579,7 @@ def print_helper(
         self,
         **kwargs: Any,
     ) -> None:
-        print(f"Submission is too big. ({format_bytes(self.size)}/{format_bytes(self.maximum_size)})")
+        print(f"Submission is too big. ({format_size(self.size)}/{format_size(self.maximum_size)})")
 
 
 class CannotParticipateException(ApiException):

crunch/cli.py

@@ -3,7 +3,7 @@
 import json
 import os
 import sys
-from typing import Any, Callable, List, Optional, Tuple, Union
+from typing import Any, Callable, List, Optional, Union
 
 import click
 

crunch/command/download.py

@@ -1,7 +1,7 @@
 import os
 import typing
 
-from .. import api, constants, container, downloader, utils
+from .. import api, constants, downloader, utils
 
 
 def _get_data_urls(
@@ -12,18 +12,14 @@ def _get_data_urls(
     int,
     int,
     typing.List[api.SplitKeyPythonType],
-    container.Features,
-    api.ColumnNames,
     typing.Dict[str, downloader.PreparedDataFile]
 ]:
     data_release = round.phases.get_submission().get_data_release(size_variant=size_variant)
 
     embargo = data_release.embargo
     number_of_features = data_release.number_of_features
-    column_names = data_release.column_names
     data_files = data_release.data_files
     splits = data_release.splits
-    features = container.Features.from_data_release(data_release)
 
     split_keys = [
         split.key
@@ -38,8 +34,6 @@ def _get_data_urls(
         embargo,
         number_of_features,
         split_keys,
-        features,
-        column_names,
         downloader.prepare_all(data_directory_path, data_files),
     )
 
@@ -87,8 +81,6 @@ def download(
         embargo,
         number_of_features,
         split_keys,
-        features,
-        column_names,
         prepared_data_files,
     ) = _get_data_urls(
         round,
@@ -111,8 +103,6 @@ def download(
         embargo,
         number_of_features,
         split_keys,
-        features,
-        column_names,
         data_directory_path,
         file_paths,
     )