app.py

import streamlit as st
# from dash import Dash, dcc, html, dash_table, Input, Output, State
import plotly.express as px
import pandas as pd
from sentence_transformers import SentenceTransformer
from scipy.spatial import distance
import re
import textwrap
import base64

question = st.text_input('Ask Marcus Aurelius a question')
dat = pd.read_csv('meditations_processed.csv')
space = pd.read_csv('sentence_embeddings.csv')

# For the update function
model_name = 'all-mpnet-base-v2'
model = SentenceTransformer(model_name)

def update_table(value):

    if not value:
        return

    text_coord = model.encode(value, show_progress_bar = True)
    out = pd.concat([dat.reset_index(), space], axis = 1)

    cos_dist = []
    embedding_cols = [str(i) for i in range(768)]
    for i in range(0, out.shape[0]):
        curr = out.iloc[i]
        curr = curr[embedding_cols]
        curr_dist = distance.cosine(u = text_coord, v = curr)
        cos_dist.append(curr_dist)

    out['cos_dist'] = cos_dist
    out = out.sort_values('cos_dist')
    out = out.head(10)
    out = out[['book', 'verse', 'text', 'cos_dist']]
    return(out)


final = update_table(question)
st.table(final)