580 mirdata multiple fixes (#581)

* Fix tox for formatting test * Pin black version to 23.1.0 * Upgrade librosa version and ensure python3.6 compatibility * Black formatting with new 23.1.0 version * Fixing egfxset expected return value * Mock pandas import at sphinx autodoc * Fix black version for python3.6
mir-dataset-loaders · Mar 7, 2023 · 0620b8c · 0620b8c
1 parent c9fd249
commit 0620b8c
Show file tree

Hide file tree

Showing 38 changed files with 15 additions and 52 deletions.
diff --git a/docs/conf.py b/docs/conf.py
@@ -46,6 +46,7 @@
     "scipy",
     "smart_open",
     "openpyxl",
+    "pandas",
 ]
 
 # # -- General configuration ---------------------------------------------------

diff --git a/mirdata/annotations.py b/mirdata/annotations.py
@@ -823,7 +823,6 @@ def _remove_duplicates(self):
             self.confidence = self.confidence[unq_idx]
 
     def __add__(self, other):
-
         if other is None:
             return self
 
@@ -1006,7 +1005,6 @@ def to_multif0(
         frequency_list: List[List[float]] = [[] for _ in times]
         confidence_list: List[List[float]] = [[] for _ in times]
         if self.confidence is not None:
-
             for t0, t1, pch, conf in zip(
                 intervals[:, 0], intervals[:, 1], self.pitches, self.confidence
             ):
@@ -1288,7 +1286,9 @@ def _from_hz(pitches_hz, target_pitch_unit):
             return pitches_midi
 
         if target_pitch_unit == "note_name":
-            return librosa.hz_to_note(pitches_hz)
+            # cast to np.array for compatibility with legacy python3.6 and
+            # librosa 0.9.2. It is redundant for librosa 0.10
+            return np.array(librosa.hz_to_note(pitches_hz))
 
         raise NotImplementedError
 

diff --git a/mirdata/datasets/billboard.py b/mirdata/datasets/billboard.py
@@ -330,7 +330,6 @@ def load_named_sections(fpath: str):
 
 
 def _load_sections(fpath: str, section_type: str):
-
     timed_sections = _parse_timed_sections(fpath)
     assert timed_sections is not None
 

diff --git a/mirdata/datasets/dagstuhl_choirset.py b/mirdata/datasets/dagstuhl_choirset.py
@@ -118,7 +118,6 @@ class Track(core.Track):
     """
 
     def __init__(self, track_id, data_home, dataset_name, index, metadata):
-
         super().__init__(
             track_id=track_id,
             data_home=data_home,
@@ -265,7 +264,6 @@ class MultiTrack(core.MultiTrack):
     def __init__(
         self, mtrack_id, data_home, dataset_name, index, track_class, metadata
     ):
-
         super().__init__(
             mtrack_id=mtrack_id,
             data_home=data_home,

diff --git a/mirdata/datasets/egfxset.py b/mirdata/datasets/egfxset.py
@@ -205,7 +205,7 @@ class Track(core.Track):
         setting (list): the setting of the effect recorded or "None" when the recording is a clean effect sound
 
     Cached Properties:
-        note_name (list): a list with the note name annotation of the audio file (e.g. "Ab5", "C6", etc.)
+        note_name (ndarray): a list with the note name annotation of the audio file (e.g. "Ab5", "C6", etc.)
         midinote (NoteData): the midinote annotation of the audio file
     """
 
@@ -337,7 +337,6 @@ def _metadata(self):
             indexname.append(name["Effect "].split(" ")[0])
 
         for track in tracknames:
-
             if track[:3] == "RAT":
                 trackiden = track[:3].lower()
 

diff --git a/mirdata/datasets/filosax.py b/mirdata/datasets/filosax.py
@@ -281,7 +281,6 @@ class Track(core.Track):
     """
 
     def __init__(self, track_id, data_home, dataset_name, index, metadata):
-
         super().__init__(
             track_id,
             data_home,

diff --git a/mirdata/datasets/orchset.py b/mirdata/datasets/orchset.py
@@ -266,7 +266,6 @@ def __init__(self, data_home=None, version="default"):
 
     @core.cached_property
     def _metadata(self):
-
         predominant_inst_path = os.path.join(
             self.data_home, "Orchset - Predominant Melodic Instruments.csv"
         )

diff --git a/mirdata/datasets/rwc_classical.py b/mirdata/datasets/rwc_classical.py
@@ -387,7 +387,6 @@ def __init__(self, data_home=None, version="default"):
 
     @core.cached_property
     def _metadata(self):
-
         metadata_path = os.path.join(self.data_home, "metadata-master", "rwc-c.csv")
 
         try:

diff --git a/mirdata/datasets/rwc_jazz.py b/mirdata/datasets/rwc_jazz.py
@@ -240,7 +240,6 @@ def __init__(self, data_home=None, version="default"):
 
     @core.cached_property
     def _metadata(self):
-
         metadata_path = os.path.join(self.data_home, "metadata-master", "rwc-j.csv")
 
         try:

diff --git a/mirdata/datasets/rwc_popular.py b/mirdata/datasets/rwc_popular.py
@@ -322,7 +322,6 @@ def __init__(self, data_home=None, version="default"):
 
     @core.cached_property
     def _metadata(self):
-
         metadata_path = os.path.join(self.data_home, "metadata-master", "rwc-p.csv")
 
         try:

diff --git a/mirdata/datasets/salami.py b/mirdata/datasets/salami.py
@@ -282,7 +282,6 @@ def __init__(self, data_home=None, version="default"):
 
     @core.cached_property
     def _metadata(self):
-
         metadata_path = os.path.join(
             self.data_home,
             os.path.join(

diff --git a/mirdata/datasets/saraga_hindustani.py b/mirdata/datasets/saraga_hindustani.py
@@ -306,7 +306,6 @@ def load_tempo(fhandle):
 
     reader = csv.reader(fhandle, delimiter=",")
     for line in reader:
-
         if "NaN" in line or " NaN" in line or "NaN " in line:
             return None
 

diff --git a/mirdata/datasets/slakh.py b/mirdata/datasets/slakh.py
@@ -121,7 +121,6 @@ class Track(core.Track):
     """
 
     def __init__(self, track_id, data_home, dataset_name, index, metadata):
-
         super().__init__(
             track_id,
             data_home,

diff --git a/mirdata/download_utils.py b/mirdata/download_utils.py
@@ -146,7 +146,6 @@ def downloader(
 
             if remotes[k].unpack_directories:
                 for src_dir in remotes[k].unpack_directories:
-
                     # path to destination directory
                     destination_dir = (
                         os.path.join(save_dir, remotes[k].destination_dir)

diff --git a/setup.py b/setup.py
@@ -31,8 +31,9 @@
         keywords="mir dataset loader audio",
         license="BSD-3-Clause",
         install_requires=[
+            "black == 22.8.0",  # last version for python3.6
             "tqdm",
-            "librosa >= 0.8.0",
+            "librosa >= 0.9.2",
             "numpy>=1.16",
             "jams",
             "requests",

diff --git a/tests/datasets/test_acousticbrainz_genre.py b/tests/datasets/test_acousticbrainz_genre.py
@@ -60,7 +60,6 @@ def test_to_jams():
 
 
 def test_filter_index():
-
     data_home = os.path.normpath("tests/resources/mir_datasets/acousticbrainz_genre")
     dataset = acousticbrainz_genre.Dataset(data_home, version="test")
     index = dataset.load_all_train()
@@ -84,7 +83,6 @@ def test_filter_index():
 
 
 def test_download(httpserver):
-
     data_home = os.path.normpath(
         "tests/resources/mir_datasets/acousticbrainz_genre_download"
     )

diff --git a/tests/datasets/test_beatles.py b/tests/datasets/test_beatles.py
@@ -61,7 +61,6 @@ def test_track():
 
 
 def test_to_jams():
-
     data_home = os.path.normpath("tests/resources/mir_datasets/beatles")
     dataset = beatles.Dataset(data_home)
     track = dataset.track("0111")

diff --git a/tests/datasets/test_billboard.py b/tests/datasets/test_billboard.py
@@ -5,7 +5,6 @@
 
 
 def test_track():
-
     default_trackid = "3"
     data_home = "tests/resources/mir_datasets/billboard"
     dataset = billboard.Dataset(data_home)
@@ -71,7 +70,6 @@ def test_track():
 
 
 def test_to_jams():
-
     default_trackid = "3"
     data_home = "tests/resources/mir_datasets/billboard"
     dataset = billboard.Dataset(data_home)

diff --git a/tests/datasets/test_dagstuhl_choirset.py b/tests/datasets/test_dagstuhl_choirset.py
@@ -166,7 +166,6 @@ def test_load_f0():
 
 
 def test_load_score():
-
     score_path = "tests/resources/mir_datasets/dagstuhl_choirset/annotations_csv_scorerepresentation/DCS_LI_QuartetB_Take04_Stereo_STM_B.csv"
     score = dagstuhl_choirset.load_score(score_path)
     assert isinstance(score, annotations.NoteData)
@@ -194,7 +193,6 @@ def test_load_score():
 
 
 def test_load_beat():
-
     beat_path = "tests/resources/mir_datasets/dagstuhl_choirset/annotations_csv_beat/DCS_LI_QuartetB_Take04_Stereo_STM.csv"
     beat = dagstuhl_choirset.load_beat(beat_path)
     assert isinstance(beat, annotations.BeatData)

diff --git a/tests/datasets/test_dali.py b/tests/datasets/test_dali.py
@@ -17,7 +17,6 @@
 
 
 def test_track():
-
     default_trackid = "4b196e6c99574dd49ad00d56e132712b"
     data_home = os.path.normpath("tests/resources/mir_datasets/dali")
     dataset = dali.Dataset(data_home)

diff --git a/tests/datasets/test_egfxset.py b/tests/datasets/test_egfxset.py
@@ -1,4 +1,5 @@
 import os
+import numpy as np
 from mirdata import annotations
 from mirdata.datasets import egfxset
 from tests.test_utils import run_track_tests
@@ -35,7 +36,7 @@ def test_track():
 
     expected_property_types = {
         "audio": tuple,
-        "note_name": list,
+        "note_name": np.ndarray,
         "midinote": annotations.NoteData,
     }
 
@@ -74,7 +75,7 @@ def test_track():
 
     expected_property_types = {
         "audio": tuple,
-        "note_name": list,
+        "note_name": np.ndarray,
         "midinote": annotations.NoteData,
     }
 
@@ -95,7 +96,6 @@ def test_track():
 
 
 def test_to_jams():
-
     data_home = os.path.normpath("tests/resources/mir_datasets/egfxset")
     dataset = egfxset.Dataset(data_home, version="test")
 

diff --git a/tests/datasets/test_ikala.py b/tests/datasets/test_ikala.py
@@ -77,7 +77,6 @@ def test_track():
 
 
 def test_to_jams():
-
     data_home = "tests/resources/mir_datasets/ikala"
     default_trackid = "10161_chorus"
     dataset = ikala.Dataset(data_home)

diff --git a/tests/datasets/test_medley_solos_db.py b/tests/datasets/test_medley_solos_db.py
@@ -31,7 +31,6 @@ def test_track():
 
 
 def test_to_jams():
-
     data_home = "tests/resources/mir_datasets/medley_solos_db"
     dataset = medley_solos_db.Dataset(data_home)
     track = dataset.track("d07b1fc0-567d-52c2-fef4-239f31c9d40e")

diff --git a/tests/datasets/test_medleydb_melody.py b/tests/datasets/test_medleydb_melody.py
@@ -53,7 +53,6 @@ def test_track():
 
 
 def test_to_jams():
-
     data_home = "tests/resources/mir_datasets/medleydb_melody"
     dataset = medleydb_melody.Dataset(data_home)
     track = dataset.track("MusicDelta_Beethoven")

diff --git a/tests/datasets/test_medleydb_pitch.py b/tests/datasets/test_medleydb_pitch.py
@@ -46,7 +46,6 @@ def test_track():
 
 
 def test_to_jams():
-
     data_home = "tests/resources/mir_datasets/medleydb_pitch"
     dataset = medleydb_pitch.Dataset(data_home)
     track = dataset.track("AClassicEducation_NightOwl_STEM_08")

diff --git a/tests/datasets/test_mtg_jamendo_autotagging_moodtheme.py b/tests/datasets/test_mtg_jamendo_autotagging_moodtheme.py
@@ -70,7 +70,6 @@ def test_to_jams():
 
 
 def test_get_track_splits():
-
     dataset = mtg_jamendo_autotagging_moodtheme.Dataset(
         "tests/resources/mir_datasets/mtg_jamendo_autotagging_moodtheme"
     )

diff --git a/tests/datasets/test_orchset.py b/tests/datasets/test_orchset.py
@@ -58,7 +58,6 @@ def test_track():
 
 
 def test_to_jams():
-
     data_home = "tests/resources/mir_datasets/orchset"
     dataset = orchset.Dataset(data_home)
     track = dataset.track("Beethoven-S3-I-ex1")

diff --git a/tests/datasets/test_rwc_classical.py b/tests/datasets/test_rwc_classical.py
@@ -51,7 +51,6 @@ def test_track():
 
 
 def test_to_jams():
-
     data_home = "tests/resources/mir_datasets/rwc_classical"
     dataset = rwc_classical.Dataset(data_home)
     track = dataset.track("RM-C003")

diff --git a/tests/datasets/test_rwc_jazz.py b/tests/datasets/test_rwc_jazz.py
@@ -5,7 +5,6 @@
 
 
 def test_track():
-
     default_trackid = "RM-J004"
     data_home = os.path.normpath("tests/resources/mir_datasets/rwc_jazz")
     dataset = rwc_jazz.Dataset(data_home)
@@ -50,7 +49,6 @@ def test_track():
 
 
 def test_to_jams():
-
     data_home = "tests/resources/mir_datasets/rwc_jazz"
     dataset = rwc_jazz.Dataset(data_home)
     track = dataset.track("RM-J004")

diff --git a/tests/datasets/test_rwc_popular.py b/tests/datasets/test_rwc_popular.py
@@ -7,7 +7,6 @@
 
 
 def test_track():
-
     default_trackid = "RM-P001"
     data_home = os.path.normpath("tests/resources/mir_datasets/rwc_popular")
     dataset = rwc_popular.Dataset(data_home)
@@ -64,7 +63,6 @@ def test_track():
 
 
 def test_to_jams():
-
     data_home = "tests/resources/mir_datasets/rwc_popular"
     dataset = rwc_popular.Dataset(data_home)
     track = dataset.track("RM-P001")

diff --git a/tests/datasets/test_salami.py b/tests/datasets/test_salami.py
@@ -6,7 +6,6 @@
 
 
 def test_track():
-
     default_trackid = "2"
     data_home = os.path.normpath("tests/resources/mir_datasets/salami")
     dataset = salami.Dataset(data_home)
@@ -122,7 +121,6 @@ def test_track():
 
 
 def test_to_jams():
-
     data_home = "tests/resources/mir_datasets/salami"
     dataset = salami.Dataset(data_home)
     track = dataset.track("2")

diff --git a/tests/datasets/test_saraga_hindustani.py b/tests/datasets/test_saraga_hindustani.py
@@ -6,7 +6,6 @@
 
 
 def test_track():
-
     default_trackid = "50_Irani_Bhairavi_Thumri"
     data_home = os.path.normpath("tests/resources/mir_datasets/saraga_hindustani")
     dataset = saraga_hindustani.Dataset(data_home)

diff --git a/tests/datasets/test_slakh.py b/tests/datasets/test_slakh.py
@@ -157,7 +157,6 @@ def test_load_tracks():
 
 
 def test_to_jams():
-
     default_trackid = "Track00001-S00"
     data_home = "tests/resources/mir_datasets/slakh"
     dataset = slakh.Dataset(data_home, version="test")
@@ -269,7 +268,6 @@ def test_multitrack():
 
 
 def test_multitrack_to_jams():
-
     default_mtrackid = "Track00001"
     data_home = "tests/resources/mir_datasets/slakh"
     dataset = slakh.Dataset(data_home, version="test")

diff --git a/tests/datasets/test_vocadito.py b/tests/datasets/test_vocadito.py
@@ -75,7 +75,6 @@ def test_track():
 
 
 def test_to_jams():
-
     data_home = "tests/resources/mir_datasets/vocadito"
     default_trackid = "1"
     dataset = vocadito.Dataset(data_home)

diff --git a/tests/resources/tmp_download_test/remote.wav b/tests/resources/tmp_download_test/remote.wav
@@ -0,0 +1 @@
+This is a test file. It is validated using checksum. Do not change the contents of this file.