def loadDataSet(): '''this loads the dataset with the genre information put inside''' ds = DataSet() try: ds.load(join(datadir, 'dataset_small.db')) except Exception as e: print('Error:', str(e)) print( 'You should first run the unittests to generate the dataset_small.db file before' ) print('trying to run this example.') sys.exit(1) genreMap = utils.getClassMapFromDirectory(join(datadir, 'dataset_small')) # transform the dataset to add the genre information ds = transform(ds, 'addfield', {'string': 'genre'}) for pid, genre in list(genreMap.items()): ds.point(pid).setLabel('genre', genre) # we're in a testing environment, prepare the dataset by cleaning it a bit more ds = transform(ds, 'removevl') # also remove the 'Sample' category, which only contains duplicates and f***s everything up... for p in ds.points(): if p['genre'] == 'Sample': ds.removePoint(p.name()) return ds
def loadDortmundDataSet(): try: MTGDB = os.environ['MTGDB_AUDIO'] except KeyError: print 'ERROR: the environment variable MTGDB_AUDIO is not defined...' print ' You should set it to point to the mount point of the MTGDB audio share.' sys.exit(1) ds = DataSet() ds.load('/tmp/dortmund.db') genreMap = utils.getClassMapFromDirectory( '%s/benchmarks/genre/Dortmund/descriptors/essentia_streaming_1.0.6_schizo' % MTGDB, sigfileExtension='.neq.sig') # transform the dataset to add the genre information ds = transform(ds, 'addfield', {'string': 'genre'}) for p in ds.points(): p['genre'] = genreMap[p.name()] # we're in a testing environment, prepare the dataset by cleaning it a bit more ds = transform(ds, 'removevl') return ds
def loadDataSet(): '''Load the dataset with the genre information put inside (as the "genre" descriptor).''' datadir = join(filedir(), '..', 'unittest', 'data') ds = DataSet() try: ds.load(join(datadir, 'dataset_small.db')) except Exception as e: print('Error:', str(e)) print( 'You should first run the unittests to generate the dataset_small.db file before' ) print('trying to run this example.') sys.exit(1) genreMap = utils.getClassMapFromDirectory(join(datadir, 'dataset_small')) # transform the dataset to add the genre information ds = transform(ds, 'addfield', {'string': 'genre'}) for p in ds.points(): p['genre'] = genreMap[p.name()] # remove the 'Sample' category, which only contains duplicates and f***s everything up... ds.collection('dataset_small').removePoints( [p.name() for p in ds.points() if p['genre'] == 'Sample']) # we're in a testing environment, prepare the dataset by cleaning it a bit more ds = transform(ds, 'removevl') return ds
def loadDortmundDataSet(): try: MTGDB = os.environ['MTGDB_AUDIO'] except KeyError: print 'ERROR: the environment variable MTGDB_AUDIO is not defined...' print ' You should set it to point to the mount point of the MTGDB audio share.' sys.exit(1) ds = DataSet() ds.load('/tmp/dortmund.db') genreMap = utils.getClassMapFromDirectory('%s/benchmarks/genre/Dortmund/descriptors/essentia_streaming_1.0.6_schizo' % MTGDB, sigfileExtension = '.neq.sig') # transform the dataset to add the genre information ds = transform(ds, 'addfield', { 'string': 'genre' }) for p in ds.points(): p['genre'] = genreMap[p.name()] # we're in a testing environment, prepare the dataset by cleaning it a bit more ds = transform(ds, 'removevl') return ds
def loadDataSet(): '''this loads the dataset with the genre information put inside''' basedir = join(filedir(), '..', 'unittest', 'data') ds = DataSet() try: ds.load(join(basedir, 'dataset_small.db')) except Exception, e: print 'Error:', str(e) print 'You should first run the unittests to generate the dataset_small.db file before' print 'trying to run this example.' sys.exit(1) genreMap = utils.getClassMapFromDirectory(join(basedir, 'dataset_small')) # transform the dataset to add the genre information ds = transform(ds, 'addfield', {'string': 'genre'}) for pid, genre in genreMap.items(): ds.point(pid).setLabel('genre', genre) # we're in a testing environment, prepare the dataset by cleaning it a bit more ds = transform(ds, 'removevl') # also remove the 'Sample' category, which only contains duplicates and f***s everything up... for p in ds.points(): if p.label('genre') == 'Sample': ds.removePoint(p.name())
def loadDataSet(): '''Load the dataset with the genre information put inside (as the "genre" descriptor).''' datadir = join(filedir(), '..', 'unittest', 'data') ds = DataSet() try: ds.load(join(datadir, 'dataset_small.db')) except Exception, e: print 'Error:', str(e) print 'You should first run the unittests to generate the dataset_small.db file before' print 'trying to run this example.' sys.exit(1) genreMap = utils.getClassMapFromDirectory(join(datadir, 'dataset_small')) # transform the dataset to add the genre information ds = transform(ds, 'addfield', { 'string': 'genre' }) for p in ds.points(): p['genre'] = genreMap[p.name()] # remove the 'Sample' category, which only contains duplicates and f***s everything up... ds.collection('dataset_small').removePoints([ p.name() for p in ds.points() if p['genre'] == 'Sample' ]) # we're in a testing environment, prepare the dataset by cleaning it a bit more ds = transform(ds, 'removevl') return ds