rename output, change default mix size (#6)

* rename output, change default mix size

* fix tests

pyproject.toml

@@ -11,7 +11,7 @@ classifiers = [
 
 [tool.poetry]
 name = "tidal_algorithmic_mixes"
-version = "0.0.4"
+version = "0.0.5"
 description = "common transformers used by the tidal personalization team."
 authors = [
     "Loay <[email protected]>",

test/discovery_mix/test_observed_tracks_aggregator.py

@@ -47,7 +47,7 @@ def test_transform(self):
         runner._data = ObservedDiscoveryMixTracksAggregatorTransformationData(observed_mixes=observed_mixes,
                                                                               mixes=mixes)
         runner.transform()
-        res = runner.output.output
+        res = runner.output.df
 
         self.assertEqual(res.columns,  [c.USER, c.TRACK_GROUP])
         self.assertEqual(res.count(), len(tracks_user_1) + len(tracks_user_2))

test/discovery_mix/test_post_processor.py

@@ -136,6 +136,6 @@ def test_transform(self):
                                                                               min_mix_size=0)
         post_processor._data = self.data
         post_processor.transform()
-        res = post_processor.output.output.collect()[0]
+        res = post_processor.output.df.collect()[0]
         self.assertEqual(Row(user=1, tracks=['xxx'], mixId='1f1451b3b417516e9e4b4423958', atDate=res.atDate),
                          res)

tidal_algorithmic_mixes/discovery_mix/daily_update_transformation.py

@@ -18,13 +18,13 @@ class DiscoveryMixDailyUpdateTransformationData:
 
 @dataclass
 class DiscoveryMixDailyUpdateTransformationOutput:
-    output: DataFrame
+    df: DataFrame
 
 
 class DiscoveryMixDailyUpdateTransformationConfig(Config):
     def __init__(self, **kwargs):
         self.current_date = kwargs.get('current_date')
-        self.mix_size = int(kwargs.get('mix_size', 70))
+        self.mix_size = int(kwargs.get('mix_size', 10))
         Config.__init__(self, **kwargs)
 
 
@@ -60,7 +60,7 @@ def transform(self, *args, **kwargs):
                                      self.config.mix_size)
                          .withColumn(c.UPDATED, F.lit(mix_utils.updated(time.time())))
                          .where(F.size(c.TRACKS) >= self.config.mix_size - 2))
-        self._output = DiscoveryMixDailyUpdateTransformationOutput(output=discovery_mix)
+        self._output = DiscoveryMixDailyUpdateTransformationOutput(discovery_mix)
 
     def slicer(self, mixes: DataFrame, current_date: date, mix_size: int) -> DataFrame:
         """ Extract the tracks of the day from the weekly computed list """

tidal_algorithmic_mixes/discovery_mix/observed_tracks_aggregator_transformation.py

@@ -18,7 +18,7 @@ class ObservedDiscoveryMixTracksAggregatorTransformationData:
 
 @dataclass
 class ObservedDiscoveryMixTracksAggregatorTransformationOutput:
-    output: DataFrame
+    df: DataFrame
 
 
 class ObservedDiscoveryMixTracksAggregatorTransformation(ETLModel):

tidal_algorithmic_mixes/discovery_mix/post_processor_transformation.py

@@ -47,7 +47,7 @@ class DiscoveryMixPostProcessorTransformationData:
 
 @dataclass
 class DiscoveryMixPostProcessorTransformationOutput:
-    output: DataFrame
+    df: DataFrame
 
 
 class DiscoveryMixPostProcessorTransformationConfig(Config):

tidal_algorithmic_mixes/discovery_mix/sasrec_model_transformation.py

@@ -2,6 +2,7 @@
 from dataclasses import dataclass
 
 from mlflow.pyfunc.spark_model_cache import SparkModelCache
+# noinspection PyProtectedMember
 from mlflow.tracking.artifact_utils import _download_artifact_from_uri
 from mlflow.utils.file_utils import TempDir
 import numpy as np
@@ -25,7 +26,7 @@ class DiscoveryMixSasRecModelTransformationData:
 
 @dataclass
 class DiscoveryMixSasRecModelTransformationOutput:
-    output: DataFrame
+    df: DataFrame
 
 
 class DiscoveryMixSasRecModelTransformationConfig(Config):