@@ -1050,33 +1050,25 @@ def grade_learner(predict, tests):
1050
1050
1051
1051
1052
1052
def train_test_split (dataset , start = None , end = None , test_split = None ):
1053
- """If you are giving 'start' and 'end' as a parameter ,
1054
- then it will return testing set from index 'start' to 'end'
1055
- and rest for training.
1056
- If you give 'test_split' as parameter then it will first shuffle the
1057
- dataset then return test_split * 100% as testing set and rest as
1053
+ """If you are giving 'start' and 'end' as parameters ,
1054
+ then it will return the testing set from index 'start' to 'end'
1055
+ and the rest for training.
1056
+ If you give 'test_split' as a parameter then it will return
1057
+ test_split * 100% as the testing set and the rest as
1058
1058
training set.
1059
1059
"""
1060
-
1061
- if start == None and end != None :
1062
- raise ValueError ("'start' parameter is missing" )
1063
-
1064
- if start != None and end == None :
1065
- raise ValueError ("'end' parameter is missing" )
1066
-
1060
+ examples = dataset .examples
1067
1061
if test_split == None :
1068
- examples = dataset .examples
1069
1062
train = examples [:start ] + examples [end :]
1070
1063
val = examples [start :end ]
1071
- return train , val
1072
1064
else :
1073
- examples = dataset .examples
1074
1065
total_size = len (examples )
1075
1066
val_size = int (total_size * test_split )
1076
1067
train_size = total_size - val_size
1077
1068
train = examples [:train_size ]
1078
1069
val = examples [train_size :total_size ]
1079
- return train , val
1070
+
1071
+ return train , val
1080
1072
1081
1073
1082
1074
def cross_validation (learner , size , dataset , k = 10 , trials = 1 ):
0 commit comments