Python help!!incorporating SMOTE using Python. Highly imbalanced dataset

[motorconcer]

I have been trying to play around with certain datasets i found on github to see how well i can conduct a sentiment analysis on different datasets and how codes work. So i have a dataset which i wanted to incorporate in the code i found the only issue is that it is a highly unbalanced dataset. for example the negative sentiment has around 5000 tweets whereas the positive has roughly 15,000 tweets. so i found different ways which i could handle this situation. The first was to use the following code using sklearn resample:

from sklearn.utils import resample
df_majority = my_df[my_df.target==1]
df_minority = my_df[my_df.target==0]

df_minority_upsampled = resample(df_minority,
replace=True,
n_samples=15025,
random_state=123)

df_upsampled = pd.concat([df_majority, df_minority_upsampled])

x = df_upsampled.Tweet

y = df_upsampled.target


from sklearn.model_selection import train_test_split

SEED = 2000

x_train, x_validation_and_test, y_train, y_validation_and_test = train_test_split(x, y, test_size=.02, random_state=SEED)

x_validation, x_test, y_validation, y_test = train_test_split(x_validation_and_test, y_validation_and_test, test_size=.5, random_state=SEED)

however using the following code i felt the results werent quite right. I then kept reading quite a bit on SMOTE which has worked extremely well with unbalanced datasets. the only issue is i have no idea how i can incorporate it into the code i found online. I am honestly really amateur at coding so some help would be appreciated. this is the following code im using:

from sklearn.feature_extraction.text import CountVectorizer
from sklearn.naive_bayes import MultinomialNB
from sklearn.linear_model import LogisticRegression
from sklearn.pipeline import Pipeline
from time import time

cvec = CountVectorizer()
lr = LogisticRegression()
n_features = np.arange(1000,20000,1000)

def nfeature_accuracy_checker(vectorizer=cvec, n_features=n_features, stop_words=None, ngram_range=(1, 1), classifier=lr):
result = []
print (classifier)
print ("\n")
for n in n_features:
vectorizer.set_params(stop_words=stop_words, max_features=n, ngram_range=ngram_range)
checker_pipeline = Pipeline([
('vectorizer', vectorizer),
('classifier', classifier)
])
print ("Validation result for {} features".format(n))
nfeature_accuracy,tt_time = accuracy_summary(checker_pipeline, x_train, y_train, x_validation, y_validation)
result.append((n,nfeature_accuracy,tt_time))
return result

my idea was to incorporate:

SMOTE_pipeline = make_pipeline(tvec, SMOTE(random_state=777),lr)

change the above code to:

def nfeature_accuracy_checker(pipeline, vectorizer=cvec, n_features=n_features, stop_words=None, ngram_range=(1, 1), classifier=lr):

then call the results using:

print ("RESULT FOR UNIGRAM WITH STOP WORDS (Tfidf)\n")
feature_result_ugt = nfeature_accuracy_checker(SMOTE_pipeline, vectorizer=tvec)

Am i thinking right or am i completely butchering the whole thing? also im happy to explain further if anyone doesnt fully understand what im trying to do. thank you