Python – Random Sample Training and Test Data from dictionary

Sometimes, while working with Machine Learning Algorithm, we can have problem in which we need to differentiate the training and testing data randomly. This is very common problem and solution to it is desirable for Machine Learning domains. This article discusses approach to solve this without using external libraries.

Method : Using keys() + random.randint() + computations
This problem can be solved by using combination of above functions. In this, we perform the task of extraction of random keys using randint(), from the keys extracted using keys(). The logical computations are performed for getting the separated test and training data.

# Python3 code to demonstrate working of
# Random Sample Training and Test Data
# Using keys() + randint() + computations
import random

# initializing dictionary
test_dict = {'gfg' : 4, 'is' : 12, 'best' : 6, 'for' : 7, 'geeks' : 10}

# printing original dictionary
print("The original dictionary is : " + str(test_dict))

# initializing ratio
test = 40
training = 60

# Random Sample Training and Test Data
# Using keys() + randint() + computations
key_list = list(test_dict.keys())

test_key_count = int((len(key_list) / 100) * test)
test_keys = [random.choice(key_list) for ele in range(test_key_count)]
train_keys = [ele for ele in key_list if ele not in test_keys]

testing_dict = dict((key, test_dict[key]) for key in test_keys
                                        if key in test_dict)
training_dict = dict((key, test_dict[key]) for key in train_keys
                                        if key in test_dict)

# printing result
print("The testing dictionary is : " + str(testing_dict))
print("The training dictionary is : " + str(training_dict))

Output :

The original dictionary is : {‘is’: 12, ‘gfg’: 4, ‘best’: 6, ‘for’: 7, ‘geeks’: 10}
The testing dictionary is : {‘is’: 12, ‘for’: 7}
The training dictionary is : {‘gfg’: 4, ‘best’: 6, ‘geeks’: 10}