app.py

import pandas as pd
import numpy as np
import seaborn as sns
import warnings
# import sklearn

# data=pd.read_csv('data.csv',error_bad_lines=False)
# data.head(5)
# data['strength'].unique()
# data.isna().sum()
# data[data['password'].isnull()]
# data.dropna(inplace=True)
# data.isnull().sum()
# sns.countplot(data['strength'])
# data.sample(10)
# password_tuple=np.array(data)
# password_tuple
# import random
# random.shuffle(password_tuple)
# x=[labels[0] for labels in password_tuple]
# y=[labels[1] for labels in password_tuple]
# def word_divide_char(inputs):
#     character=[]
#     for i in inputs:
#         character.append(i)
#     return character
# word_divide_char('kzde5577')
# from sklearn.feature_extraction.text import TfidfVectorizer
# vectorizer=TfidfVectorizer(tokenizer=word_divide_char)
# X=vectorizer.fit_transform(x)
# X.shape
# vectorizer.get_feature_names_out()
# first_document_vector=X[0]
# first_document_vector
# first_document_vector.T.todense()
# df=pd.DataFrame(first_document_vector.T.todense(),index=vectorizer.get_feature_names_out(),columns=['TF-IDF'])
# df.sort_values(by=['TF-IDF'],ascending=False)
import joblib

model = joblib.load('finalized_model.sav')

new_data = 'sdhb%jksdn&73e4d';
new_data2=np.array([new_data])
new_data3=vectorizer.transform(new_data2)
predicted = model.predict(new_data3)
print(predicted)