Add script to download midi-files ot midi folder

download_midi.py

@@ -0,0 +1,56 @@
+__author__ = 'val'
+
+
+from BeautifulSoup import BeautifulSoup
+import urllib2
+import re
+import os
+
+dict = set()
+
+def download_midi_recursive(website, page, folder):
+    if page in dict:
+        return
+
+    dict.add(page)
+    print "Downloading page " + page
+
+    html_page = urllib2.urlopen(website + '/' + page)
+    soup = BeautifulSoup(html_page)
+    for link in soup.findAll('a'):
+        url = '{}'.format(link.get('href'))
+
+        if url.endswith('.mid'):
+            try:
+                filename = os.path.basename(url)
+                midiurl = urllib2.urlopen(website + '/' + url)
+                fullpath = folder + '/' + filename
+
+                if os.path.exists(fullpath):
+                    print "Skipping " + filename
+                else:
+                    print "Downloading " +  filename
+                    with open(fullpath, "wb") as local_file:
+                        content = midiurl.read()
+                        local_file.write(content)
+
+            except urllib2.HTTPError, e:
+                print "Http error" + e.code + url
+            except urllib2.URLError, e:
+                print "Url error" + e.reason + url
+        if url.endswith('.htm'):
+            try:
+                relativeurl = os.path.basename(url)
+                download_midi_recursive(website, relativeurl, folder)
+            except Exception, e:
+                print e.message
+
+
+
+
+# website = "http://www.midiworld.com"
+website = "http://www.piano-midi.de"
+# page = "classic.htm"
+page = "midi_files.htm"
+folder = './music'
+download_midi_recursive(website, page, folder)

main.py

@@ -1,4 +1,5 @@
 import cPickle as pickle
+import os.path
 import gzip
 import numpy
 from midi_to_statematrix import *
@@ -37,12 +38,21 @@ def fetch_train_thoughts(m,pcs,batches,name="trainthoughts"):
 		all_thoughts.append((ipt,opt,thoughts))
 	pickle.dump(all_thoughts, open('output/'+name+'.p','wb'))
 
+def create_output_directory():
+	outputdir = "output"
+	try:
+		os.stat(outputdir)
+	except:
+		os.mkdir(outputdir)
+
 if __name__ == '__main__':
 
 	pcs = multi_training.loadPieces("music")
 
 	m = model.Model([300,300],[100,50], dropout=0.5)
 
+	create_output_directory()
+
 	multi_training.trainPiece(m, pcs, 10000)
 
 	pickle.dump( m.learned_config, open( "output/final_learned_config.p", "wb" ) )

multi_training.py

@@ -19,12 +19,15 @@ def loadPieces(dirpath):
 
         name = fname[:-4]
 
-        outMatrix = midiToNoteStateMatrix(os.path.join(dirpath, fname))
-        if len(outMatrix) < batch_len:
-            continue
-
-        pieces[name] = outMatrix
-        print "Loaded {}".format(name)
+        try:
+            outMatrix = midiToNoteStateMatrix(os.path.join(dirpath, fname))
+            if len(outMatrix) < batch_len:
+                continue
+
+            pieces[name] = outMatrix
+            print "Loaded {}".format(name)
+        except Exception, e:
+            print "Skipping to load {} due to error".format(name)
 
     return pieces