Question 4 0.5 pts Joe writes the following code to separate MNIST data into testing, training, and validation sets. The output of his code is also shown. What values he has used for A= B= C= and D- ? Note: The answers must be in numbers and in x.x format. You could refer to the lecture to see the required format of your response. from keran.dataneta import mist import matplotlib.pyplot as plt import numpy np (*_train all, ytrain 11), x_test, y_test) -mist.load_data) num_train_img_train all shape 01 train_Indrap.range(0, num_train_img train_ind_sup.randon.permutation train_ind) print(train_ind.shape) *_train_shuttledex_train_alltrain_ind_.. y train shuttledwy, tenin alltrain_and_ #selecting training data *_traine_train_shuttled int(num_trans.1.01 y trainy train hurled int(Bestrating selecting remaining as the validation data *_wab-x_trai_shuffled[interm_training). y_subay_tralashuffledorint num_trai_ing) peint the number of images in training set tsx_train shape[]) print("The number of images in validation at 18.b.shapatos print the number of images in testing at test baper (60000.0 The mber of images in training anti 48000 The mber of images in validation set is 12000 The number of ages in testing set 1 10000 Question 4 0.5 pts Joe writes the following code to separate MNIST data into testing, training, and validation sets. The output of his code is also shown. What values he has used for A= B= C= and D- ? Note: The answers must be in numbers and in x.x format. You could refer to the lecture to see the required format of your response. from keran.dataneta import mist import matplotlib.pyplot as plt import numpy np (*_train all, ytrain 11), x_test, y_test) -mist.load_data) num_train_img_train all shape 01 train_Indrap.range(0, num_train_img train_ind_sup.randon.permutation train_ind) print(train_ind.shape) *_train_shuttledex_train_alltrain_ind_.. y train shuttledwy, tenin alltrain_and_ #selecting training data *_traine_train_shuttled int(num_trans.1.01 y trainy train hurled int(Bestrating selecting remaining as the validation data *_wab-x_trai_shuffled[interm_training). y_subay_tralashuffledorint num_trai_ing) peint the number of images in training set tsx_train shape[]) print("The number of images in validation at 18.b.shapatos print the number of images in testing at test baper (60000.0 The mber of images in training anti 48000 The mber of images in validation set is 12000 The number of ages in testing set 1 10000