Move feature extraction to separate folder, Init batch_feature_extractor

classifier/README.md → classifier/feature_extraction/README.md

@@ -1,16 +1,20 @@
 # Feature extraction
 
-The file `feature_extractor` is a python module that uses the open-source library [Essentia](http://essentia.upf.edu/documentation/index.html) to extract audio features from a file in the path specified in the first parameter and save the features' values to a binary file in the path specified in the second parameter.
+The file `feature_extractor` is a python module that uses the open-source library [Essentia](http://essentia.upf.edu/documentation/index.html) to extract audio features from an audio file in the path specified in the first parameter and save the features' values to a json file in the path specified in the second parameter.
+
+The module can be imported or executed as a script using one of the following commands
+`python feature_extractor.py <audio_file_path> <extracted_features_file_path> <audio_file_sample_rate>`
+or
+`python3 feature_extractor.py <audio_file_path> <extracted_features_file_path> <audio_file_sample_rate>`
+
+A python script is also provided for a batch feature extraction. The script can be executed using one of the following commands:
+`python batch_feature_extractor.py <audio_files_directory> <feature_files_directory> <audio_files_sample_rate>`
+or
+`python3 batch_feature_extractor.py <audio_files_directory> <feature_files_directory> <audio_files_sample_rate>`
 
 **Dependencies:**
 - essentia
 - numpy
 - scipy
-- matplotlib
-
-All dependencies are available both for python2 and python3 versions and can all be installed using the commands `pip install <package_name>` or `pip3 install <package_name>` for python2 and python3 respectively.
 
-The module can be imported or executed as a script using one of the following commands
+All dependencies are available both for python2 and python3 versions and can all be installed using the commands `pip install <package_name>` or `pip3 install <package_name>` for python2 and python3 respectively.
-`python feature_extractor.py <audio_file_path> <extracted_features_file_path> <audio_file_sample_rate>`
-or
-`python3 feature_extractor.py <audio_file_path> <extracted_features_file_path> <audio_file_sample_rate>`

classifier/feature_extraction/batch_feature_extractor.py

@@ -0,0 +1,17 @@
+import sys
+from os import listdir
+from os.path import isfile, join
+import multiprocessing as mp
+from feature_extractor import extractFeatures
+
+audioFiles = [file for file in listdir(sys.argv[1]) if isfile(join(sys.argv[1], file))]
+
+# Without multithreading
+# for file in audioFiles:
+# 	extractFeatures(sys.argv[1] + file,
+# 		sys.argv[2] + file[0:file.rfind('.')] + '.json', int(sys.argv[3]))
+
+pool = mp.Pool(processes = 8)
+[pool.apply(extractFeatures, args=(sys.argv[1] + file,
+		sys.argv[2] + file[0:file.rfind('.')] + '.json',
+		int(sys.argv[3]))) for file in audioFiles]

classifier/feature_extractor.py → classifier/feature_extraction/feature_extractor.py

@@ -1,13 +1,15 @@
+# import essentia.standard
 import essentia
-import essentia.standard
+from essentia.standard import (MonoLoader, Windowing, Spectrum, MFCC,
-from essentia.standard import *
+	ZeroCrossingRate, SpectralCentroidTime, RollOff, Flux, FrameGenerator,
-import essentia.streaming
+	YamlOutput)
-from pylab import plot, show, figure, imshow
+
-import matplotlib.pyplot as plt
+# Disable annoying info level logging
+essentia.log.infoActive = False
 
 def extractFeatures(audioPath, outputPath, sampleRate):
 	# Loads the audio file specified
-	loader = essentia.standard.MonoLoader(filename = audioPath, sampleRate = sampleRate)
+	loader = MonoLoader(filename = audioPath, sampleRate = sampleRate)
 	audio = loader()
 
 	# Sets up the functions that will be used