Statsbomb: even more fixes

README.md

@@ -42,6 +42,7 @@ from kloppy import (
     load_metrica_tracking_data, 
     load_tracab_tracking_data,
     load_epts_tracking_data, 
+    load_statsbomb_event_data,
     to_pandas, 
     transform
 )
@@ -53,6 +54,10 @@ dataset = load_tracab_tracking_data('meta.xml', 'raw_data.txt')
 # or epts
 dataset = load_epts_tracking_data('meta.xml', 'raw_data.txt')
 
+# or event data
+dataset = load_statsbomb_event_data('event_data.json', 'lineup.json')
+
+
 dataset = transform(dataset, pitch_dimensions=[[0, 108], [-34, 34]])
 pandas_data_frame = to_pandas(dataset)
 ```
@@ -143,6 +148,30 @@ with open("raw_data.txt", "rb") as raw, \
 ```
 
 
+or StatsBomb event data
+```python
+from kloppy import StatsBombSerializer
+
+serializer = StatsBombSerializer()
+
+with open("events/123123.json", "rb") as event_data, \
+        open("lineup/123123.json", "rb") as lineup_data:
+
+    dataset = serializer.deserialize(
+        inputs={
+            'event_data': event_data,
+            'lineup_data': lineup_data
+        },
+        options={
+            "event_types": ["pass", "shot", "carry", "take_on"]
+        }
+    )
+
+    # start working with dataset
+```
+
+
+
 ### <a name="pitch-dimensions"></a>Transform the pitch dimensions
 Data providers use their own pitch dimensions. Some use actual meters while others use 100x100. Use the Transformer to get from one pitch dimensions to another one.
 ```python
@@ -195,7 +224,7 @@ Data models
 - [ ] Automated tests
 - [x] Pitch
 - [x] Tracking
-- [ ] Event
+- [x] Event
 
 Tracking data (de)serializers
 - [x] Automated tests
@@ -207,7 +236,7 @@ Tracking data (de)serializers
 Event data (de)serializers
 - [ ] Automated tests
 - [ ] OPTA
-- [ ] StatsBomb
+- [x] StatsBomb
 - [ ] MetricaSports
 
 Transformers

examples/datasets/metrica.py

@@ -1,3 +1,6 @@
+import logging
+import sys
+
 from kloppy import datasets, to_pandas
 
 
@@ -6,6 +9,9 @@ def main():
         This example shows the use of Metrica datasets, and how we can pass argument
         to the dataset loader.
     """
+    logging.basicConfig(stream=sys.stdout, level=logging.INFO,
+                        format="%(asctime)s - %(name)s - %(levelname)s - %(message)s")
+
 
     # The metrica dataset loader loads by default the 'game1' dataset
     dataset = datasets.load("metrica_tracking", options={'sample_rate': 1./12, 'limit': 10})

examples/datasets/statsbomb.py

@@ -1,7 +1,7 @@
 import logging
 import sys
 
-from kloppy import datasets, transform, to_pandas
+from kloppy import datasets, transform, to_pandas, load_statsbomb_event_data
 from kloppy.infra.utils import performance_logging
 
 
@@ -32,6 +32,12 @@ def main():
 
     print(dataframe[:100].to_string())
 
+    # or load it using the helper from disk
+    dataset = load_statsbomb_event_data(
+        "events/15946.json",
+        "lineups/15946.json"
+    )
+
 
 if __name__ == "__main__":
     main()

examples/epts/load_epts_into_pandas.py

@@ -1,3 +1,6 @@
+import logging
+import sys
+
 from pandas import DataFrame
 
 from kloppy.infra.serializers.tracking.epts.meta_data import load_meta_data as epts_load_meta_data
@@ -17,6 +20,9 @@ def main():
     4. Try to consume items from generator twice
     4. Convert the records into a pandas dataframe for easy data mangling
     """
+    logging.basicConfig(stream=sys.stdout, level=logging.INFO,
+                        format="%(asctime)s - %(name)s - %(levelname)s - %(message)s")
+
 
     # step 1: load metadata
     with open("epts_meta.xml", "rb") as meta_fp:

examples/playing_time.py

@@ -1,3 +1,5 @@
+import logging
+import sys
 from collections import Counter
 
 from kloppy import datasets
@@ -8,6 +10,8 @@ def main():
     """
         This example shows how to determine playing time
     """
+    logging.basicConfig(stream=sys.stdout, level=logging.INFO,
+                        format="%(asctime)s - %(name)s - %(levelname)s - %(message)s")
 
     dataset = datasets.load("metrica_tracking", options={'sample_rate': 1./25})
 

kloppy/helpers.py

@@ -49,15 +49,15 @@ def load_epts_tracking_data(meta_data_filename: str, raw_data_filename: str, opt
         )
 
 
-def load_statsbomb_event_data(lineup_filename: str, raw_data_filename: str, options: dict = None) -> EventDataset:
+def load_statsbomb_event_data(event_data_filename: str, lineup_data_filename: str, options: dict = None) -> EventDataset:
     serializer = StatsBombSerializer()
-    with open(lineup_filename, "rb") as lineup_data, \
-            open(raw_data_filename, "rb") as raw_data:
+    with open(event_data_filename, "rb") as event_data, \
+            open(lineup_data_filename, "rb") as lineup_data:
 
         return serializer.deserialize(
             inputs={
-                'lineup_data': lineup_data,
-                'raw_data': raw_data
+                'event_data': event_data,
+                'lineup_data': lineup_data
             },
             options=options
         )
@@ -166,6 +166,7 @@ def to_pandas(dataset: Dataset, _record_converter: Callable = None) -> 'DataFram
     'load_tracab_tracking_data',
     'load_metrica_tracking_data',
     'load_epts_tracking_data',
+    'load_statsbomb_event_data',
     'to_pandas',
     'transform'
 ]

kloppy/infra/serializers/event/statsbomb/serializer.py

@@ -217,6 +217,8 @@ def deserialize(self, inputs: Dict[str, Readable], options: Dict = None) -> Even
                 >>>     )
                 """
         self.__validate_inputs(inputs)
+        if not options:
+            options = {}
 
         with performance_logging("load data", logger=logger):
             raw_events = json.load(inputs['event_data'])

kloppy/infra/utils.py

@@ -24,7 +24,7 @@ def performance_logging(description: str, counter: int = None, logger=None):
             extra = f" ({int(counter / took * 1000)}items/sec)"
 
         unit = "ms"
-        if took < 0.01:
+        if took < 0.1:
             took *= 1000
             unit = "us"