In [1]:

    

import numpy as np
from sklearn.model_selection import train_test_split


    

In [2]:

    

a = np.arange(10)
print(a)


    

    




[0 1 2 3 4 5 6 7 8 9]

In [3]:

    

print(train_test_split(a))


    

    




[array([3, 9, 6, 1, 5, 0, 7]), array([2, 8, 4])]

In [4]:

    

print(type(train_test_split(a)))


    

    




<class 'list'>

In [5]:

    

print(len(train_test_split(a)))


    

    




2

In [6]:

    

a_train, a_test = train_test_split(a)


    

In [7]:

    

print(a_train)


    

    




[3 4 0 5 7 8 2]

In [8]:

    

print(a_test)


    

    




[6 1 9]

In [9]:

    

a_train, a_test = train_test_split(a, test_size=0.6)


    

In [10]:

    

print(a_train)


    

    




[9 1 2 6]

In [11]:

    

print(a_test)


    

    




[5 7 4 3 0 8]

In [12]:

    

a_train, a_test = train_test_split(a, test_size=6)


    

In [13]:

    

print(a_train)


    

    




[4 2 1 0]

In [14]:

    

print(a_test)


    

    




[7 6 3 9 8 5]

In [15]:

    

a_train, a_test = train_test_split(a, train_size=0.6)


    

In [16]:

    

print(a_train)


    

    




[2 9 6 0 4 3]

In [17]:

    

print(a_test)


    

    




[7 8 5 1]

In [18]:

    

a_train, a_test = train_test_split(a, train_size=6)


    

In [19]:

    

print(a_train)


    

    




[9 3 0 8 7 1]

In [20]:

    

print(a_test)


    

    




[5 6 4 2]

In [21]:

    

a_train, a_test = train_test_split(a, train_size=0.25)


    

In [22]:

    

print(a_train)


    

    




[1 2]

In [23]:

    

print(a_test)


    

    




[0 8 4 7 5 6 3 9]

In [24]:

    

a_train, a_test = train_test_split(a, test_size=0.3, train_size=0.4)


    

In [25]:

    

print(a_train)


    

    




[3 0 4 9]

In [26]:

    

print(a_test)


    

    




[7 2 8]

In [27]:

    

a_train, a_test = train_test_split(a, test_size=3, train_size=4)


    

In [28]:

    

print(a_train)


    

    




[9 7 0 4]

In [29]:

    

print(a_test)


    

    




[3 8 5]

In [30]:

    

# a_train, a_test = train_test_split(a, test_size=0.8, train_size=0.7)
# ValueError: The sum of test_size and train_size = 1.500000, should be smaller than 1.0. Reduce test_size and/or train_size.


    

In [31]:

    

# a_train, a_test = train_test_split(a, test_size=8, train_size=7)
# ValueError: The sum of train_size and test_size = 15, should be smaller than the number of samples 10. Reduce test_size and/or train_size.


    

In [32]:

    

a_train, a_test = train_test_split(a, shuffle=False)


    

In [33]:

    

print(a_train)


    

    




[0 1 2 3 4 5 6]

In [34]:

    

print(a_test)


    

    




[7 8 9]

In [35]:

    

a_train, a_test = train_test_split(a, random_state=0)


    

In [36]:

    

print(a_train)


    

    




[9 1 6 7 3 0 5]

In [37]:

    

print(a_test)


    

    




[2 8 4]

In [38]:

    

X = np.arange(20).reshape(2, 10).T
print(X)


    

    




[[ 0 10]
 [ 1 11]
 [ 2 12]
 [ 3 13]
 [ 4 14]
 [ 5 15]
 [ 6 16]
 [ 7 17]
 [ 8 18]
 [ 9 19]]

In [39]:

    

y = np.arange(10)
print(y)


    

    




[0 1 2 3 4 5 6 7 8 9]

In [40]:

    

X_train, X_test, y_train, y_test = train_test_split(X, y)


    

In [41]:

    

print(X_train)


    

    




[[ 7 17]
 [ 3 13]
 [ 0 10]
 [ 8 18]
 [ 6 16]
 [ 4 14]
 [ 2 12]]

In [42]:

    

print(X_test)


    

    




[[ 5 15]
 [ 1 11]
 [ 9 19]]

In [43]:

    

print(y_train)


    

    




[7 3 0 8 6 4 2]

In [44]:

    

print(y_test)


    

    




[5 1 9]

In [45]:

    

z = np.arange(10) * 10
print(z)


    

    




[ 0 10 20 30 40 50 60 70 80 90]

In [46]:

    

X_train, X_test, y_train, y_test, z_train, z_test = train_test_split(X, y, z)


    

In [47]:

    

print(X_train)


    

    




[[ 6 16]
 [ 9 19]
 [ 1 11]
 [ 2 12]
 [ 7 17]
 [ 0 10]
 [ 3 13]]

In [48]:

    

print(X_test)


    

    




[[ 8 18]
 [ 4 14]
 [ 5 15]]

In [49]:

    

print(y_train)


    

    




[6 9 1 2 7 0 3]

In [50]:

    

print(y_test)


    

    




[8 4 5]

In [51]:

    

print(z_train)


    

    




[60 90 10 20 70  0 30]

In [52]:

    

print(z_test)


    

    




[80 40 50]

In [53]:

    

y_mismatch = np.arange(8)
print(y_mismatch)


    

    




[0 1 2 3 4 5 6 7]

In [54]:

    

# X_train, X_test, y_train, y_test = train_test_split(X, y_mismatch)
# ValueError: Found input variables with inconsistent numbers of samples: [10, 8]


    

In [55]:

    

y = np.array([0] * 5 + [1] * 5)
print(y)


    

    




[0 0 0 0 0 1 1 1 1 1]

In [56]:

    

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=100)


    

In [57]:

    

print(y_train)


    

    




[0 1 0 0 0 0 1 1]

In [58]:

    

print(y_test)


    

    




[1 1]

In [59]:

    

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=100,
                                                    stratify=y)


    

In [60]:

    

print(y_train)


    

    




[1 1 0 0 0 1 1 0]

In [61]:

    

print(y_test)


    

    




[1 0]