# -*- coding: utf-8 -*-
"""
Created on Mon Nov 24 08:47:17 2025

@author: ispan
"""

import numpy as np;
import matplotlib.pyplot as plt;
from sklearn.datasets import fetch_20newsgroups;
from sklearn.feature_extraction.text import CountVectorizer;
from sklearn.decomposition import LatentDirichletAllocation as LDA;

categories = [
    'sci.space',
    'rec.autos',
    'talk.politics.guns'
];

ds_train = fetch_20newsgroups(subset='train',categories=categories); # data_home='C:/scikit_learn_data',
ds_test = fetch_20newsgroups(subset='test',categories=categories);
n_train = len(ds_train.data);
n_test = len(ds_test.data);
        
        
analyze = CountVectorizer(stop_words='english',max_df=0.8,min_df=0.01); 
DT_train = analyze.fit_transform(ds_train.data); 
vocabulary = analyze.vocabulary_; 
words = analyze.get_feature_names_out();
n_words = DT_train.shape[1];

myLDA = LDA(n_components=3);
myLDA.fit(DT_train);

for i in range(n_words):
    print(words[i],myLDA.components_[:,i]);
    
doc_prob_train = myLDA.transform(DT_train);
score_train = myLDA.score(DT_train);
    
    