pelase finsih this in accordance with MNIST dataset

class gen_classifier(object):

def __init__(self, x_train, x_test, y_train, y_test):

self.x_train=x_train

self.x_test=x_test

self.y_train=y_train

self.y_test=y_test

self.seed = 1000 # For reproducibility

self.unique_classes= np.unique(y_train)

def fit_decision_tree_classifier(self):

#####

# Add code here to train/fit decision tree on mnist

clf_dt=None

return clf_dt

def fit_random_forest_classifier(self):

#####

# Add code here to train/fit random forest on mnist

clf_rf=None

return clf_rf

def clf_predict(self,trained_clf=None):

### Use your trained classifier to make predictions

### Replace following hard coded lines with your code

y_pred_train = np.zeros(self.y_train.shape)

y_pred_test = np.zeros(self.y_test.shape)

####

return y_pred_train, y_pred_test, self.y_train, self.y_test

def per_class_metrics(self, y_true, y_pred, classifier_name=None, split=None):

TP=[]

TN=[]

Accuracy=[]

Precision=[]

Recall=[]

F1score=[]

Label=[]

for label in self.unique_classes:

# Write code here that calculates the following for current label and assigns the values in variable named = (tp,tn,acc,prec,recall, f1)

###################

# Your code here

# For example, hard coded to 0:

tp = 0

tn= 0

acc = 0

prec = 0

rec = 0

f1= 0

##########################

TP.append(tp)

TN.append(tn)

Accuracy.append(acc)

Precision.append(prec)

Recall.append(rec)

F1score.append(f1)

Label.append(label)

clf_name=classifier_name

text=''

for idx in range(len(Label)):

text += 'For class : {}, Following are the metrics with {} on {} set True Positive : {} True Negative : {} Accuracy : {} Precision : {} Recall : {} F1score : {} '.format(Label[idx],clf_name, split ,TP[idx],TN[idx],Accuracy[idx],Precision[idx],Recall[idx],F1score[idx])

print(text)

def overall_metrics(self, y_true, y_pred,classifier_name=None, split=None):

# Write code here that calculates the following and assigns the values in variable named = (tp,tn,acc,prec,recall, f1)

###################

# Your code here

# For example, hard coded to 0:

tp = 0

tn= 0

acc = 0

prec = 0

rec = 0

f1= 0

##########################

clf_name=classifier_name

text = 'Following are the metrics for {} on {} set: True Positive : {} True Negative : {} Accuracy : {} Precision : {} Recall : {} F1score : {} '.format(clf_name, split ,tp,tn,acc,prec,rec,f1)

print(text)