import pandas as pd
import tiktoken
import os
import openai
from openai.embeddings_utils import get_embedding, cosine_similarity
import numpy as np
import streamlit as st 

input_datapath = "fine_food_reviews_with_embeddings_1k.csv"
df = pd.read_csv(input_datapath, index_col=0)

st.title("Semanti Search")


#adding another column having the summary as title and the actual text as content
df["combined"] = (
    "Title: " + df.Summary.str.strip() + "; Content: " + df.Text.str.strip()
)


# embedding model parameters
embedding_model = "text-embedding-ada-002"
embedding_encoding = "cl100k_base"  # this the encoding for text-embedding-ada-002
max_tokens = 8000  # the maximum for text-embedding-ada-002 is 8191


encoding = tiktoken.get_encoding(embedding_encoding)
top_n = 500
# omit reviews that are too long to embed
df["n_tokens"] = df.combined.apply(lambda x: len(encoding.encode(x)))
df = df[df.n_tokens <= max_tokens].tail(top_n)


datafile_path = "fine_food_reviews_with_embeddings_1k.csv"
df = pd.read_csv(datafile_path)
df["embedding"] = df.embedding.apply(eval).apply(np.array)