Merge pull request #25 from bmcfee/unlabeled-transform

Better syntax for unlabeled transformation

pumpp/core.py

@@ -16,18 +16,20 @@
 from .feature import FeatureExtractor
 
 
-def transform(audio_f, jams_f, *ops):
+def transform(audio_f, *ops, **kwargs):
     '''Apply a set of operations to a track
 
     Parameters
     ----------
     audio_f : str
         The path to the audio file
 
-    jams_f : str
-        The path to the jams file
+    jam : str, jams.JAMS, or file-like
+        A JAMS object, or path to a JAMS file.
 
-    ops : list of pumpp.task.BaseTaskTransform or pumpp.feature.FeatureExtractor
+        If not provided, an empty jams object will be created.
+
+    ops : list of task.BaseTaskTransform or feature.FeatureExtractor
         The operators to apply to the input data
 
     Returns
@@ -39,8 +41,15 @@ def transform(audio_f, jams_f, *ops):
     # Load the audio
     y, sr = librosa.load(audio_f, sr=None, mono=True)
 
+    jam = kwargs.pop('jam', None)
+
+    if jam is None:
+        jam = jams.JAMS()
+        jam.file_metadata.duration = librosa.get_duration(y=y, sr=sr)
+
     # Load the jams
-    jam = jams.load(jams_f)
+    if not isinstance(jam, jams.JAMS):
+        jam = jams.load(jam)
 
     data = dict()
 

tests/test_core.py

@@ -6,37 +6,49 @@
 import numpy as np
 
 import pumpp
+import jams
 
 
 @pytest.fixture(params=[11025, 22050])
 def sr(request):
     return request.param
 
 
-
 @pytest.fixture(params=[128, 512])
 def hop_length(request):
     return request.param
 
 
-@pytest.mark.parametrize('audio_f, jams_f',
-                         [('tests/data/test.ogg',
-                           'tests/data/test.jams')])
-def test_transform(audio_f, jams_f, sr, hop_length):
+@pytest.fixture(params=[None,
+                        'tests/data/test.jams',
+                        jams.load('tests/data/test.jams')])
+def jam(request):
+    return request.param
+
+
+@pytest.mark.parametrize('audio_f', ['tests/data/test.ogg'])
+def test_transform(audio_f, jam, sr, hop_length):
 
     ops = [pumpp.feature.STFT(name='stft', sr=sr,
                               hop_length=hop_length,
                               n_fft=2*hop_length),
-           pumpp.task.BeatTransformer(name='beat', sr=sr, hop_length=hop_length),
-           pumpp.task.ChordTransformer(name='chord', sr=sr, hop_length=hop_length),
-           pumpp.task.StaticLabelTransformer(name='tags', namespace='tag_open',
+
+           pumpp.task.BeatTransformer(name='beat', sr=sr,
+                                      hop_length=hop_length),
+
+           pumpp.task.ChordTransformer(name='chord', sr=sr,
+                                       hop_length=hop_length),
+
+           pumpp.task.StaticLabelTransformer(name='tags',
+                                             namespace='tag_open',
                                              labels=['rock', 'jazz'])]
 
-    data = pumpp.transform(audio_f, jams_f, *ops)
+    data = pumpp.transform(audio_f, jam=jam, *ops)
 
     # Fields we should have:
     assert set(data.keys()) == set(['stft/mag', 'stft/phase',
-                                    'beat/beat', 'beat/downbeat', 'beat/_valid',
+                                    'beat/beat', 'beat/downbeat',
+                                    'beat/_valid',
                                     'beat/mask_downbeat',
                                     'chord/pitch', 'chord/root', 'chord/bass',
                                     'chord/_valid',
@@ -48,7 +60,7 @@ def test_transform(audio_f, jams_f, sr, hop_length):
     assert data['beat/beat'].shape[1] == data['chord/root'].shape[1]
     assert data['beat/beat'].shape[1] == data['chord/bass'].shape[1]
 
-    # Audio features can be off by 
-    assert (np.abs(data['stft/mag'].shape[1] - data['beat/beat'].shape[1]) 
+    # Audio features can be off by at most a frame
+    assert (np.abs(data['stft/mag'].shape[1] - data['beat/beat'].shape[1])
             * hop_length / float(sr)) <= 0.05
     pass