Merge pull request #12 from ivd-git/naming-consistency

Naming consistency

.gitignore

@@ -0,0 +1,2 @@
+.idea/
+__pycache__/

README.md

@@ -47,14 +47,14 @@ from kloppy import (
 )
 
 # metrica data
-data_set = load_metrica_tracking_data('home_file.csv', 'away_file.csv')
+dataset = load_metrica_tracking_data('home_file.csv', 'away_file.csv')
 # or tracab
-data_set = load_tracab_tracking_data('meta.xml', 'raw_data.txt')
+dataset = load_tracab_tracking_data('meta.xml', 'raw_data.txt')
 # or epts
-data_set = load_epts_tracking_data('meta.xml', 'raw_data.txt')
+dataset = load_epts_tracking_data('meta.xml', 'raw_data.txt')
 
-data_set = transform(data_set, pitch_dimensions=[[0, 108], [-34, 34]])
-pandas_data_frame = to_pandas(data_set)
+dataset = transform(dataset, pitch_dimensions=[[0, 108], [-34, 34]])
+pandas_data_frame = to_pandas(dataset)
 ```
 
 ### <a name="datasets"></a>Public datasets / Very quick start
@@ -63,7 +63,7 @@ we added a "dataset loader" which does all the heavy lifting for you: find urls,
 ```python
 from kloppy import datasets
 
-data_set = datasets.load("metrica_tracking", options={'sample_rate': 1./12, 'limit': 10})
+dataset = datasets.load("metrica_tracking", options={'sample_rate': 1./12, 'limit': 10})
 ```
 
 ### <a name="models"></a>Standardized models
@@ -84,7 +84,7 @@ serializer = TRACABSerializer()
 with open("tracab_data.dat", "rb") as raw, \
         open("tracab_metadata.xml", "rb") as meta:
 
-    data_set = serializer.deserialize(
+    dataset = serializer.deserialize(
         inputs={
             'raw_data': raw,
             'meta_data': meta
@@ -94,7 +94,7 @@ with open("tracab_data.dat", "rb") as raw, \
         }
     )
 
-    # start working with data_set
+    # start working with dataset
 ```
 
 or Metrica data
@@ -106,7 +106,7 @@ serializer = MetricaTrackingSerializer()
 with open("Sample_Game_1_RawTrackingData_Away_Team.csv", "rb") as raw_away, \
         open("Sample_Game_1_RawTrackingData_Home_Team.csv", "rb") as raw_home:
 
-    data_set = serializer.deserialize(
+    dataset = serializer.deserialize(
         inputs={
             'raw_data_home': raw_home,
             'raw_data_away': raw_away
@@ -116,7 +116,7 @@ with open("Sample_Game_1_RawTrackingData_Away_Team.csv", "rb") as raw_away, \
         }
     )
 
-    # start working with data_set
+    # start working with dataset
 ```
 
 
@@ -129,7 +129,7 @@ serializer = EPTSSerializer()
 with open("raw_data.txt", "rb") as raw, \
         open("metadata.xml", "rb") as meta:
 
-    data_set = serializer.deserialize(
+    dataset = serializer.deserialize(
         inputs={
             'raw_data': raw,
             'meta_data': meta
@@ -139,7 +139,7 @@ with open("raw_data.txt", "rb") as raw, \
         }
     )
 
-    # start working with data_set
+    # start working with dataset
 ```
 
 
@@ -148,9 +148,9 @@ Data providers use their own pitch dimensions. Some use actual meters while othe
 ```python
 from kloppy.domain import Transformer, PitchDimensions, Dimension
 
-# use deserialized `data_set`
-new_data_set = Transformer.transform_data_set(
-    data_set,
+# use deserialized `dataset`
+new_dataset = Transformer.transform_dataset(
+    dataset,
     to_pitch_dimensions=PitchDimensions(
         x_dim=Dimension(0, 100),
         y_dim=Dimension(0, 100)
@@ -166,8 +166,8 @@ Data providers can use different orientations. Some use a fixed orientation and
 ```python
 from kloppy.domain import Transformer, Orientation
 
-new_data_set = Transformer.transform_data_set(
-    data_set,
+new_dataset = Transformer.transform_dataset(
+    dataset,
     to_orientation=Orientation.BALL_OWNING_TEAM
 )
 ```
@@ -176,9 +176,9 @@ new_data_set = Transformer.transform_data_set(
 ```python
 from kloppy.domain import Transformer, PitchDimensions, Dimension, Orientation
 
-# use deserialized `data_set`
-new_data_set = Transformer.transform_data_set(
-    data_set,
+# use deserialized `dataset`
+new_dataset = Transformer.transform_dataset(
+    dataset,
     to_pitch_dimensions=PitchDimensions(
         x_dim=Dimension(0, 100),
         y_dim=Dimension(0, 100)

examples/datasets/metrica.py

@@ -8,13 +8,13 @@ def main():
     """
 
     # The metrica dataset loader loads by default the 'game1' dataset
-    data_set = datasets.load("metrica_tracking", options={'sample_rate': 1./12, 'limit': 10})
-    print(len(data_set.frames))
+    dataset = datasets.load("metrica_tracking", options={'sample_rate': 1./12, 'limit': 10})
+    print(len(dataset.frames))
 
     # We can pass additional keyword arguments to the loaders to specify a different dataset
-    data_set = datasets.load("metrica_tracking", options={'limit': 1000}, game='game2')
+    dataset = datasets.load("metrica_tracking", options={'limit': 1000}, game='game2')
 
-    data_frame = to_pandas(data_set)
+    data_frame = to_pandas(dataset)
     print(data_frame)
 
 

examples/playing_time.py

@@ -9,10 +9,10 @@ def main():
         This example shows how to determine playing time
     """
 
-    data_set = datasets.load("metrica_tracking", options={'sample_rate': 1./25})
+    dataset = datasets.load("metrica_tracking", options={'sample_rate': 1./25})
 
     playing_seconds_per_player = Counter()
-    for frame in data_set.frames:
+    for frame in dataset.frames:
         playing_seconds_per_player.update([int(jersey_no) for jersey_no in frame.home_team_player_positions.keys()])
 
     x = range(len(playing_seconds_per_player))

kloppy/domain/models/common.py

@@ -93,7 +93,7 @@ def set_attacking_direction(self, attacking_direction: AttackingDirection):
         self.attacking_direction = attacking_direction
 
 
-class DataSetFlag(Flag):
+class DatasetFlag(Flag):
     BALL_OWNING_TEAM = 1
     BALL_STATE = 2
 
@@ -108,8 +108,8 @@ class DataRecord(ABC):
 
 
 @dataclass
-class DataSet(ABC):
-    flags: DataSetFlag
+class Dataset(ABC):
+    flags: DatasetFlag
     pitch_dimensions: PitchDimensions
     orientation: Orientation
     periods: List[Period]

kloppy/domain/models/event.py

@@ -6,7 +6,7 @@
 from typing import List, Union
 
 from .pitch import Point
-from .common import DataRecord, DataSet, Team
+from .common import DataRecord, Dataset, Team
 
 
 class SubType(Enum):
@@ -226,7 +226,7 @@ def event_type(self) -> EventType:
 
 
 @dataclass
-class EventDataSet(DataSet):
+class EventDataset(Dataset):
     records: List[Union[
         SetPieceEvent, ShotEvent
     ]]

kloppy/domain/models/tracking.py

@@ -2,7 +2,7 @@
 from typing import List, Dict
 
 from .common import (
-    DataSet,
+    Dataset,
     DataRecord
 )
 from .pitch import Point
@@ -17,7 +17,7 @@ class Frame(DataRecord):
 
 
 @dataclass
-class TrackingDataSet(DataSet):
+class TrackingDataset(Dataset):
     frame_rate: int
     records: List[Frame]
 

kloppy/domain/services/enrichers/__init__.py

@@ -1,12 +1,12 @@
 from dataclasses import dataclass
 
-from ...models.tracking import DataSet as TrackingDataSet
-from ...models.event import DataSet as EventDataSet
-from ...models.common import DataSetFlag, Team, BallState
+from ...models.tracking import Dataset as TrackingDataset
+from ...models.event import Dataset as EventDataset
+from ...models.common import DatasetFlag, Team, BallState
 
 
 class TrackingPossessionEnricher:
-    def enrich_inplace(self, tracking_data_set: TrackingDataSet, event_data_set: EventDataSet) -> None:
+    def enrich_inplace(self, tracking_dataset: TrackingDataset, event_dataset: EventDataset) -> None:
         """
             Return an enriched tracking data set.
 
@@ -16,20 +16,20 @@ def enrich_inplace(self, tracking_data_set: TrackingDataSet, event_data_set: Eve
             they happen.
 
         """
-        if tracking_data_set.flags & (DataSetFlag.BALL_OWNING_TEAM | DataSetFlag.BALL_STATE):
+        if tracking_dataset.flags & (DatasetFlag.BALL_OWNING_TEAM | DatasetFlag.BALL_STATE):
             return
 
-        if not event_data_set.flags & (DataSetFlag.BALL_OWNING_TEAM | DataSetFlag.BALL_STATE):
-            raise Exception("Event DataSet does not contain ball owning team or ball state information")
+        if not event_dataset.flags & (DatasetFlag.BALL_OWNING_TEAM | DatasetFlag.BALL_STATE):
+            raise Exception("Event Dataset does not contain ball owning team or ball state information")
 
         # set some defaults
         next_event_idx = 0
         current_ball_owning_team = None
         current_ball_state = None
 
-        for frame in tracking_data_set.records:
-            if next_event_idx < len(event_data_set.records):
-                event = event_data_set.records[next_event_idx]
+        for frame in tracking_dataset.records:
+            if next_event_idx < len(event_dataset.records):
+                event = event_dataset.records[next_event_idx]
                 if frame.period.id == event.period.id and frame.timestamp >= event.timestamp:
                     current_ball_owning_team = event.ball_owning_team
                     current_ball_state = event.ball_state

kloppy/domain/services/transformers/__init__.py

@@ -7,7 +7,7 @@
     Frame,
     Team, AttackingDirection,
 
-    TrackingDataSet, DataSetFlag, DataSet, # NOT YET: EventDataSet
+    TrackingDataset, DatasetFlag, Dataset, # NOT YET: EventDataset
 )
 
 
@@ -78,43 +78,43 @@ def transform_frame(self, frame: Frame) -> Frame:
             }
         )
 
-    DataSetType = TypeVar('DataSetType')
+    DatasetType = TypeVar('DatasetType')
 
     @classmethod
-    def transform_data_set(cls,
-                           data_set: DataSetType,
+    def transform_dataset(cls,
+                           dataset: DatasetType,
                            to_pitch_dimensions: PitchDimensions = None,
-                           to_orientation: Orientation = None) -> DataSetType:
+                           to_orientation: Orientation = None) -> DatasetType:
         if not to_pitch_dimensions and not to_orientation:
-            return data_set
+            return dataset
         elif not to_orientation:
-            to_orientation = data_set.orientation
+            to_orientation = dataset.orientation
         elif not to_pitch_dimensions:
-            to_pitch_dimensions = data_set.pitch_dimensions
+            to_pitch_dimensions = dataset.pitch_dimensions
 
         if to_orientation == Orientation.BALL_OWNING_TEAM:
-            if not data_set.flags & DataSetFlag.BALL_OWNING_TEAM:
+            if not dataset.flags & DatasetFlag.BALL_OWNING_TEAM:
                 raise ValueError("Cannot transform to BALL_OWNING_TEAM orientation when dataset doesn't contain "
                                  "ball owning team data")
 
         transformer = cls(
-            from_pitch_dimensions=data_set.pitch_dimensions,
-            from_orientation=data_set.orientation,
+            from_pitch_dimensions=dataset.pitch_dimensions,
+            from_orientation=dataset.orientation,
             to_pitch_dimensions=to_pitch_dimensions,
             to_orientation=to_orientation
         )
-        if isinstance(data_set, TrackingDataSet):
-            frames = list(map(transformer.transform_frame, data_set.records))
+        if isinstance(dataset, TrackingDataset):
+            frames = list(map(transformer.transform_frame, dataset.records))
 
-            return TrackingDataSet(
-                flags=data_set.flags,
-                frame_rate=data_set.frame_rate,
-                periods=data_set.periods,
+            return TrackingDataset(
+                flags=dataset.flags,
+                frame_rate=dataset.frame_rate,
+                periods=dataset.periods,
                 pitch_dimensions=to_pitch_dimensions,
                 orientation=to_orientation,
                 records=frames
             )
-        #elif isinstance(data_set, EventDataSet):
-        #    raise Exception("EventDataSet transformer not implemented yet")
+        #elif isinstance(dataset, EventDataset):
+        #    raise Exception("EventDataset transformer not implemented yet")
         else:
-            raise Exception("Unknown DataSet type")
+            raise Exception("Unknown Dataset type")