import pandas as pd
import topicwizard
from sklearn.decomposition import NMF
from topicwizard.pipeline import make_topic_pipeline
from sklearn.feature_extraction.text import CountVectorizer

df = pd.read_csv('df_merged.csv')
abstracts=df['description'].tolist()
vectorizer = CountVectorizer(min_df=5, max_df=0.8, stop_words="english")
model = NMF(n_components=10)
topic_pipeline = make_topic_pipeline(vectorizer, model)
topic_pipeline.fit(abstracts)
topicwizard.visualize(abstracts, pipeline=topic_pipeline)

app = topicwizard.get_dash_app(vectorizer, model, corpus=abstracts)

# main.py
if __name__ == "__main__":
    app.run_server(debug=False, port=7860)