scikt-learn

csv2arff-read-only/.svn/all-wcprops

+K 25
+svn:wc:ra_dav:version-url
+V 21
+/svn/!svn/ver/3/trunk
+END
+csv2arff.py
+K 25
+svn:wc:ra_dav:version-url
+V 33
+/svn/!svn/ver/2/trunk/csv2arff.py
+END

csv2arff-read-only/.svn/entries

+10
+
+dir
+3
+http://csv2arff.googlecode.com/svn/trunk
+http://csv2arff.googlecode.com/svn
+
+
+
+2008-07-17T09:36:53.100851Z
+3
+bianchimro
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+6882ac45-3452-0410-b3ff-f72446878f82
+
+csv2arff.py
+file
+
+
+
+
+2013-06-20T08:20:15.000000Z
+3cd48889501c569b6bd45ba00a39be7d
+2008-07-17T09:32:31.216611Z
+2
+bianchimro
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+3968
+
+test_csv2arff
+dir
+

csv2arff-read-only/.svn/text-base/csv2arff.py.svn-base

+# -*- coding: cp1252 -*-
+import csv
+import sys
+from xml.dom import minidom
+
+
+
+
+
+
+def get_attributes(file_xml):
+    out = []
+    dom1 = minidom.parse(file_xml)
+    for node in dom1.getElementsByTagName('attribute'):
+        out.append({
+            'name': node.getAttribute('name') ,
+            'atype': node.getAttribute('atype'),
+            'format':node.getAttribute('format'),
+            'skip':node.getAttribute('skip')
+
+        })
+    #print out
+    return out
+
+def get_relation(file_xml):
+    dom1 = minidom.parse(file_xml)
+    out=''
+    delimiter=''
+    for node in dom1.getElementsByTagName('csv'):
+        out=node.getAttribute('name')
+        delimiter=node.getAttribute('delimiter');
+    if(len(delimiter)==0):
+        delimiter=';';
+    print delimiter   
+    return out, delimiter
+
+
+class csv_arff_converter:
+    
+    def __init__(self,csv_file, attribute_file, file_out):
+        self.csv_file = csv_file
+        self.attribute_file = attribute_file
+        self.file_out = file_out
+
+    def run(self):
+
+        classes = []
+        
+        #read attribute
+        self.relation_name, self.delimiter = get_relation(attribute_file)
+        attributes_list = get_attributes(attribute_file)
+        arff_data = '@RELATION ' + self.relation_name + '\n\n'
+         
+
+        for i in attributes_list:
+            if (i['skip'] != 'yes'):
+                arff_data += '@ATTRIBUTE '+i['name']+' ' + i['atype']
+                if (i['atype']=='date'):
+                    arff_data += ' '+i['format']
+                if (i['atype']=='class'):
+                    arff_data += ' (#@#'+i['name'] + '#@#)'
+                
+                arff_data +='\n'
+                classes.append('')
+             
+
+
+        arff_data += '\n@DATA\n'
+        print classes 
+        #open csv
+        reader = csv.reader(open(self.csv_file), delimiter=self.delimiter, quoting=csv.QUOTE_NONE)
+
+        rnum = 0     
+
+
+        for row in reader:
+            #print row
+            buff = ''
+            pos = 0
+            #print len(row)
+            #occhio alla lunghezza riga
+            for j in range(0, len(row)-1):
+                field = row[j]
+                
+                if(attributes_list[pos]['skip'] != 'yes'):
+                
+                    if (pos > 0):
+                        buff += ','
+                    if(attributes_list[pos]['atype'] == 'string'):
+                        field = "'" + field + "'"
+                    buff += field
+                    #se  una classe raccolgo i valori
+                    if(attributes_list[pos]['atype'] == 'class'):
+                        if (rnum > 0):
+                            classes[pos]+= ','+ field
+                        else:
+                            classes[pos]+=  field
+                        
+                pos += 1
+            buff += '\n'
+            arff_data += buff
+            
+            rnum += 1
+
+        
+        pos = 0
+        for a in classes:
+            j = a.split(',')
+            
+            un = list(set(j))
+            #print un
+            if (len(un) > 0):
+                this_replacement =  ",".join(un)
+                #print this_replacement
+                old_text = '#@#'+ attributes_list[pos]['name'] + '#@#'
+                #print old_text
+                arff_data = arff_data.replace(old_text, this_replacement)
+            pos += 1
+
+        #print arff_data
+        a = open(self.file_out, 'w')
+        a.write(arff_data)
+        a.close()
+        
+                
+                
+
+
+
+if __name__ == "__main__":
+    #csv_file = sys.argv[1]
+    #attribute_file = sys.argv[2]
+    csv_file = './test_csv2arff/test_dataset_1.csv'
+    attribute_file = './test_csv2arff/test_dataset_1.att'
+        
+    instance = csv_arff_converter(csv_file, attribute_file, './test_csv2arff/output.arff')
+    instance.run()

csv2arff-read-only/csv2arff.py

+# -*- coding: cp1252 -*-
+import csv
+import sys
+from xml.dom import minidom
+
+
+
+
+
+
+def get_attributes(file_xml):
+    out = []
+    dom1 = minidom.parse(file_xml)
+    for node in dom1.getElementsByTagName('attribute'):
+        out.append({
+            'name': node.getAttribute('name') ,
+            'atype': node.getAttribute('atype'),
+            'format':node.getAttribute('format'),
+            'skip':node.getAttribute('skip')
+
+        })
+    #print out
+    return out
+
+def get_relation(file_xml):
+    dom1 = minidom.parse(file_xml)
+    out=''
+    delimiter=''
+    for node in dom1.getElementsByTagName('csv'):
+        out=node.getAttribute('name')
+        delimiter=node.getAttribute('delimiter');
+    if(len(delimiter)==0):
+        delimiter=';';
+    print delimiter   
+    return out, delimiter
+
+
+class csv_arff_converter:
+    
+    def __init__(self,csv_file, attribute_file, file_out):
+        self.csv_file = csv_file
+        self.attribute_file = attribute_file
+        self.file_out = file_out
+
+    def run(self):
+
+        classes = []
+        
+        #read attribute
+        self.relation_name, self.delimiter = get_relation(attribute_file)
+        attributes_list = get_attributes(attribute_file)
+        arff_data = '@RELATION ' + self.relation_name + '\n\n'
+         
+
+        for i in attributes_list:
+            if (i['skip'] != 'yes'):
+                arff_data += '@ATTRIBUTE '+i['name']+' ' + i['atype']
+                if (i['atype']=='date'):
+                    arff_data += ' '+i['format']
+                if (i['atype']=='class'):
+                    arff_data += ' (#@#'+i['name'] + '#@#)'
+                
+                arff_data +='\n'
+                classes.append('')
+             
+
+
+        arff_data += '\n@DATA\n'
+        print classes 
+        #open csv
+        reader = csv.reader(open(self.csv_file), delimiter=self.delimiter, quoting=csv.QUOTE_NONE)
+
+        rnum = 0     
+
+
+        for row in reader:
+            #print row
+            buff = ''
+            pos = 0
+            #print len(row)
+            #occhio alla lunghezza riga
+            for j in range(0, len(row)-1):
+                field = row[j]
+                
+                if(attributes_list[pos]['skip'] != 'yes'):
+                
+                    if (pos > 0):
+                        buff += ','
+                    if(attributes_list[pos]['atype'] == 'string'):
+                        field = "'" + field + "'"
+                    buff += field
+                    #se  una classe raccolgo i valori
+                    if(attributes_list[pos]['atype'] == 'class'):
+                        if (rnum > 0):
+                            classes[pos]+= ','+ field
+                        else:
+                            classes[pos]+=  field
+                        
+                pos += 1
+            buff += '\n'
+            arff_data += buff
+            
+            rnum += 1
+
+        
+        pos = 0
+        for a in classes:
+            j = a.split(',')
+            
+            un = list(set(j))
+            #print un
+            if (len(un) > 0):
+                this_replacement =  ",".join(un)
+                #print this_replacement
+                old_text = '#@#'+ attributes_list[pos]['name'] + '#@#'
+                #print old_text
+                arff_data = arff_data.replace(old_text, this_replacement)
+            pos += 1
+
+        #print arff_data
+        a = open(self.file_out, 'w')
+        a.write(arff_data)
+        a.close()
+        
+                
+                
+
+
+
+if __name__ == "__main__":
+    #csv_file = sys.argv[1]
+    #attribute_file = sys.argv[2]
+    csv_file = './test_csv2arff/test_dataset_1.csv'
+    attribute_file = './test_csv2arff/test_dataset_1.att'
+        
+    instance = csv_arff_converter(csv_file, attribute_file, './test_csv2arff/output.arff')
+    instance.run()

csv2arff-read-only/test_csv2arff/.svn/all-wcprops

+K 25
+svn:wc:ra_dav:version-url
+V 35
+/svn/!svn/ver/3/trunk/test_csv2arff
+END
+test_dataset_1.att
+K 25
+svn:wc:ra_dav:version-url
+V 54
+/svn/!svn/ver/3/trunk/test_csv2arff/test_dataset_1.att
+END
+output.arff
+K 25
+svn:wc:ra_dav:version-url
+V 47
+/svn/!svn/ver/3/trunk/test_csv2arff/output.arff
+END
+test_dataset_1.csv
+K 25
+svn:wc:ra_dav:version-url
+V 54
+/svn/!svn/ver/3/trunk/test_csv2arff/test_dataset_1.csv
+END

csv2arff-read-only/test_csv2arff/.svn/entries

+10
+
+dir
+3
+http://csv2arff.googlecode.com/svn/trunk/test_csv2arff
+http://csv2arff.googlecode.com/svn
+
+
+
+2008-07-17T09:36:53.100851Z
+3
+bianchimro
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+6882ac45-3452-0410-b3ff-f72446878f82
+
+test_dataset_1.att
+file
+
+
+
+
+2013-06-20T08:20:15.000000Z
+bbc1c0f68459e4837932dd23adc9d878
+2008-07-17T09:36:53.100851Z
+3
+bianchimro
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+1717
+
+output.arff
+file
+
+
+
+
+2013-06-20T08:20:15.000000Z
+e13d8ccc4be16ce6f650fd61edc00d9a
+2008-07-17T09:36:53.100851Z
+3
+bianchimro
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+1030624
+
+test_dataset_1.csv
+file
+
+
+
+
+2013-06-20T08:20:15.000000Z
+7b517897ae99c70aed31b21030a7ed3e
+2008-07-17T09:36:53.100851Z
+3
+bianchimro
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+1947734
+

csv2arff-read-only/test_csv2arff/.svn/text-base/test_dataset_1.att.svn-base

+<?xml version="1.0"?>
+<csv name="dataset_1" delimiter=";">
+	<attribute name="dataora"  atype="string" skip="no"/>
+	<attribute name="eur_usd_aggregates_future_60_diff_massimo"  atype="numeric"/>
+	<attribute name="eur_usd_aggregates_past_30_close_avg"  atype="numeric"/>
+	<attribute name="eur_usd_aggregates_past_30_close_var"  atype="numeric"  skip="yes"/>
+	<attribute name="eur_usd_aggregates_past_300_close_avg"  atype="numeric"/>
+	<attribute name="eur_usd_aggregates_past_300_close_var"  atype="numeric"  skip="yes"/>
+	<attribute name="eur_usd_aggregates_past_1200_close_avg"  atype="numeric"/>
+	<attribute name="eur_usd_aggregates_past_1200_close_var"  atype="numeric" skip="yes"/>
+	<attribute name="eur_gbp_aggregates_past_30_close_avg"  atype="numeric"/>
+	<attribute name="eur_gbp_aggregates_past_30_close_var"  atype="numeric"  skip="yes"/>
+	<attribute name="eur_gbp_aggregates_past_300_close_avg"  atype="numeric" />
+	<attribute name="eur_gbp_aggregates_past_300_close_var"  atype="numeric"  skip="yes"/>
+	<attribute name="eur_gbp_aggregates_past_1200_close_avg"  atype="numeric"/>
+	<attribute name="eur_gbp_aggregates_past_1200_close_var"  atype="numeric"  skip="yes"/>
+	<attribute name="gbp_usd_aggregates_past_30_close_avg"  atype="numeric"/>
+	<attribute name="gbp_usd_aggregates_past_30_close_var"  atype="numeric"  skip="yes"/>
+	<attribute name="gbp_usd_aggregates_past_300_close_avg"  atype="numeric"/>
+	<attribute name="gbp_usd_aggregates_past_300_close_var"  atype="numeric"  skip="yes"/>
+	<attribute name="gbp_usd_aggregates_past_1200_close_avg"  atype="numeric"/>
+	<attribute name="gbp_usd_aggregates_past_1200_close_var"  atype="numeric"  skip="yes"/>
+ 
+	
+	
+</csv>
+
+
+ 
\ No newline at end of file

csv2arff-read-only/test_csv2arff/test_dataset_1.att

+<?xml version="1.0"?>
+<csv name="dataset_1" delimiter=";">
+	<attribute name="dataora"  atype="string" skip="no"/>
+	<attribute name="eur_usd_aggregates_future_60_diff_massimo"  atype="numeric"/>
+	<attribute name="eur_usd_aggregates_past_30_close_avg"  atype="numeric"/>
+	<attribute name="eur_usd_aggregates_past_30_close_var"  atype="numeric"  skip="yes"/>
+	<attribute name="eur_usd_aggregates_past_300_close_avg"  atype="numeric"/>
+	<attribute name="eur_usd_aggregates_past_300_close_var"  atype="numeric"  skip="yes"/>
+	<attribute name="eur_usd_aggregates_past_1200_close_avg"  atype="numeric"/>
+	<attribute name="eur_usd_aggregates_past_1200_close_var"  atype="numeric" skip="yes"/>
+	<attribute name="eur_gbp_aggregates_past_30_close_avg"  atype="numeric"/>
+	<attribute name="eur_gbp_aggregates_past_30_close_var"  atype="numeric"  skip="yes"/>
+	<attribute name="eur_gbp_aggregates_past_300_close_avg"  atype="numeric" />
+	<attribute name="eur_gbp_aggregates_past_300_close_var"  atype="numeric"  skip="yes"/>
+	<attribute name="eur_gbp_aggregates_past_1200_close_avg"  atype="numeric"/>
+	<attribute name="eur_gbp_aggregates_past_1200_close_var"  atype="numeric"  skip="yes"/>
+	<attribute name="gbp_usd_aggregates_past_30_close_avg"  atype="numeric"/>
+	<attribute name="gbp_usd_aggregates_past_30_close_var"  atype="numeric"  skip="yes"/>
+	<attribute name="gbp_usd_aggregates_past_300_close_avg"  atype="numeric"/>
+	<attribute name="gbp_usd_aggregates_past_300_close_var"  atype="numeric"  skip="yes"/>
+	<attribute name="gbp_usd_aggregates_past_1200_close_avg"  atype="numeric"/>
+	<attribute name="gbp_usd_aggregates_past_1200_close_var"  atype="numeric"  skip="yes"/>
+ 
+	
+	
+</csv>
+
+
+ 
\ No newline at end of file

mothra/settings.py

@@ -199,6 +199,7 @@ INSTALLED_APPS_WORKFLOWS_SUB = (
     'workflows.mysql',
     'workflows.lemmagen',
     'workflows.crossbee',
+    'workflows.scikitAlgorithms',
     #WORKFLOWS_SUBAPP_PLACEHOLDER
     )
 

requirements.txt

@@ -10,4 +10,10 @@ httplib2==0.7.5
 pyparsing==1.5.6
 pydot==1.0.28
 wsgiref==0.1.2
-mysql-connector-python==1.0.9
\ No newline at end of file
+mysql-connector-python==1.0.9
+
+# scikit-learn requires
+# NumPy>=1.3