changeset 8:28f15e232028

test to get audio samples URL from 7 digital

--- a/Code/read_taste_profile.py	Sun Jul 12 23:56:25 2015 +0100
+++ b/Code/read_taste_profile.py	Tue Jul 14 23:41:55 2015 +0100
@@ -2,33 +2,74 @@
 import csv
 import pandas as pd
 import numpy as np
+import itertools
+import time
 
 # List of h5 files (audio streams)
-with open('/homes/pchilguano/dataset/cal10kHDF5.csv', 'wb') as out:
+#with open('/homes/pchilguano/dataset/cal10kHDF5.csv', 'wb') as out:
+#	writer = csv.writer(out, delimiter=',')
+#	for root, dirs, files in os.walk("/homes/pchilguano/dataset/cal10k"):
+#		for file in files:
+#			if file.endswith(".h5"):
+#				#print(os.path.join(root, file))
+#				track = ''.join(['SO',str(file)[2:-3]])
+#				print(track)
+#				writer.writerow([track])
+				
+#with open('/homes/pchilguano/dataset/cal10kHDF5.csv', 'rb') as f:
+#	reader = csv.reader(f)
+#	your_list = list(reader)
+#	your_list.sort()
+#	chain = list(itertools.chain(*your_list))
+
+
+#store = pd.HDFStore('/homes/pchilguano/dataset/store.h5')
+location = r'~/dataset/train_triplets.txt'
+chunksize = 10000
+count = 0
+frame = pd.DataFrame()
+for chunk in pd.read_csv(location, delim_whitespace=True, header=None, names=['user','song','plays'], chunksize=chunksize):
+	#chunk.sort(columns='song')
+	#frame = chunk.query('song == your_list')
+	frame = frame.append(chunk.pivot(index='user', columns='song', values='plays'), ignore_index=True)
+	count = count + 1
+	print(count)
+	#for item in your_list:
+		#chunk['song'].isin(item)
+	#store.append('df', chunk[chunk['song'].isin(item)])
+#store.close()
+	
+	
+df = pd.read_csv(location, delim_whitespace=True, header=None, names=['user','song','plays'])
+ddf = df.drop_duplicates(subset = 'song')
+ddf.to_csv('train_triplets_song.csv',columns=['song'], header=False, index=False)
+
+with open('/homes/pchilguano/dataset/sid_mismatches.txt', 'rb') as f, open('/homes/pchilguano/dataset/sid_mismatches_song.txt', 'wb') as out:
 	writer = csv.writer(out, delimiter=',')
-	for root, dirs, files in os.walk("/homes/pchilguano/dataset/cal10k"):
-		for file in files:
-			if file.endswith(".h5"):
-				#print(os.path.join(root, file))
-				track = ''.join(['SO',str(file)[2:-3]])
-				print(track)
-				writer.writerow([track])
-				
-with open('/homes/pchilguano/dataset/cal10kHDF5.csv', 'rb') as f:
-    reader = csv.reader(f)
-    your_list = list(reader)
+	next = f.readline()
+	while next != "":
+		writer.writerow([next[8:26]])
+		print(next[8:26])
+		next = f.readline()
+#mismatch.to_csv('sid_mismatches_song.csv',columns=1, header=False, index=False)
 
 
-store = pd.HDFStore('/homes/pchilguano/dataset/store.h5')
-location = r'~/dataset/train_triplets.txt'
-chunksize = 4
-for chunk in pd.read_csv(location, delim_whitespace=True, header=None, names=['user','song','plays'], chunksize=chunksize):
-	#frame = pd.Dataframe()
-	#frame = chunk.query('song == your_list')
-	frame = chunk.pivot(index='user', columns='song', values='plays')
-	store.append('df', frame)
-
-	
+from pyechonest import song, config
+config.ECHO_NEST_API_KEY="LINDFDUTQZQ781IE8"
+with open('/homes/pchilguano/dataset/test_echonest.txt', 'rb') as input, open('/homes/pchilguano/dataset/test_echonest_url.txt', 'wb') as output:
+	writer = csv.writer(output, delimiter=',')
+	next = input.readline()
+	while next != "":
+		time.sleep(1)
+		s = song.Song(next[:-2])
+		time.sleep(1)
+		ss_tracks = s.get_tracks('7digital-UK')
+		if len(ss_tracks) != 0:
+			ss_track = ss_tracks[0]
+			preview_url = ss_track.get('preview_url')
+			print(preview_url)
+			writer.writerow([next[:-2], preview_url])
+		next = input.readline()
 	
 	
 df = store['df']