Multinominal Naive Bayes Works with one example but not the other (sciki-learn)

Question

I'm just doing a super simple BBC news classifier to learn. I found this one code example online and it worked (it predicts if a town is in the countryside or by the city):

data = [{'house': 100, 'street': 50, 'shop': 25, 'car': 100, 'tree': 20},
{'house': 5, 'street': 5, 'shop': 0, 'car': 10, 'tree': 500, 'river': 1}]

dv = DictVectorizer(sparse=False)
X = dv.fit_transform(data)
Y = np.array([1, 0])

mnb = MultinomialNB()
mnb.fit(X, Y)

test_data = data = [{'house': 80, 'street': 20, 'shop': 15, 'car': 70, 'tree': 10, 'river': 
1}, {'house': 10, 'street': 5, 'shop': 1, 'car': 8, 'tree': 300, 'river': 0}]

mnb.predict(dv.fit_transform(test_data))

So I take this and I apply it to BBC I found here with this code here:

df_train = pd.read_csv("BBC-News-Train.csv")
df_test = pd.read_csv("bbc_train_jack.csv")

def clean_text(text):
    # remove everything except alphabets
    text = re.sub("[^a-zA-Z]", " ", text)
    # remove whitespaces
    text = ' '.join(text.split())
    text = text.lower()
    
    return text

df_train['clean_text'] = df_train['Text'].apply(clean_text)
df_test['clean_text'] = df_test['Text'].apply(clean_text)

stop_words = set(stopwords.words('english'))

# function to remove stopwords
def remove_stopwords(text):
    no_stopword_text = [w for w in text.split() if not w in stop_words]
    return no_stopword_text

df_train['clean_text'] = df_train['clean_text'].apply(lambda x: remove_stopwords(x))
df_test['clean_text'] = df_test['clean_text'].apply(lambda x: remove_stopwords(x))

def get_id(x):
    y = 0
    if x == 'business':
        y = 0
    if x == 'entertainment':
        y = 1
    if x == 'tech':
        y = 2
    if x == 'politics':
        y = 3
    if x == 'sport':
        y = 4
    return y

df_train['category_id'] = df_train['Category'].apply(get_id)
df_test['category_id'] = df_test['Category'].apply(get_id)

def makeWordCount(listOfWords):
    counts = Counter(listOfWords)
    return dict(counts.most_common(5))

df_train['word_count'] = df_train['clean_text'].apply(makeWordCount)
df_test['word_count'] = df_test['clean_text'].apply(makeWordCount)

def getLen(d):
    return len(d)

df_train['len_word_count'] = df_train['word_count'].apply(getLen)
df_test['len_word_count'] = df_test['word_count'].apply(getLen)

print(df_test)

data = df_train['word_count'].to_list()

print(data[:10])
print("")

dv = DictVectorizer(sparse=False)
X = dv.fit_transform(data)
y = np.array(df_train['category_id'].to_list())

clf = MultinomialNB()
clf.fit(X, y)

test_data = data = [{'eviction': 6, 'said': 4, 'show': 13, 'chart': 9, 'radio': 7}]

clf.predict(dv.fit_transform(test_data))

But I'm getting this error:

ValueError: X has 5 features, but MultinomialNB is expecting 2121 features as input.

Why is this happening? The test is me just dividing the files into two different ones