# HG changeset patch
# User Maria Panteli
# Date 1507200446 -3600
# Node ID 7a8d043d2d45c7b17e7c6a76c5c69a09fbbb91be
# Parent  d6c937bc15313ebb035253a377194b32da4e0e1a
more on metadata for public code

diff -r d6c937bc1531 -r 7a8d043d2d45 scripts/add_collection.py
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/scripts/add_collection.py	Thu Oct 05 11:47:26 2017 +0100
@@ -0,0 +1,21 @@
+import numpy as np
+import pandas as pd
+import pickle 
+import os
+
+
+METADATA_FILE = '../../music-outliers/data/metadata.csv'
+#METADATA_FILE = '../data/metadata.csv'
+
+df = pd.read_csv(METADATA_FILE)
+
+collection = []
+for audio in df['Audio']:
+    if 'SampleAudio' in audio:
+        collection.append('Smithsonian Folkways')
+    else:
+        collection.append('British Library')
+collection = np.array(collection)
+df['Collection'] = collection
+
+df.to_csv('../../music-outliers/data/metadata_2.csv', index=False)