awacke1/Wikipedia.Chat.Multiplayer

1import streamlit as st2import spacy3import wikipediaapi4import wikipedia5from wikipedia.exceptions import DisambiguationError6from transformers import TFAutoModel, AutoTokenizer7import numpy as np8import pandas as pd9import faiss10import datetime11import time12 13 14st.markdown("""15 16Scene 1: The Enchanted Castle17 18You arrive at the enchanted castle, surrounded by a forest of thorns. You have heard stories of a beautiful princess asleep within, waiting for someone to awaken her.19 20Option 1: Try to make your way through the thorns.21Option 2: Look for another way in.22 23Sentiment: Feels like harsher trials after passive sleep.24 25---26 27Scene 2: The Castle's Secrets28 29If you made it past the thorns, you discover that the castle is full of hidden chambers, each containing a different trial. 30 31These trials are designed to test your limits, reveal your inner most desires, and help you understand the suffering of humankind.32 33Option 1: Enter the first chamber.34Option 2: Continue exploring the castle.35 36Sentiment: Comedy ending in marriage.37 38---39 40Scene 3: The Princess's Awakening41 42After navigating the castle's trials, you finally reach the chamber where the princess lies sleeping. 43 44You are faced with the decision of how to awaken her, knowing that your actions will determine the nature of your relationship with her.45 46Option 1: Awaken her with a gentle kiss.47Option 2: Awaken her through a more assertive act like lifting her up.48 49Sentiment: Heart forged awakening with different implications depending on context.50 51""")52 53try:54    nlp = spacy.load("en_core_web_sm")55except:56    spacy.cli.download("en_core_web_sm")57    nlp = spacy.load("en_core_web_sm")58 59wh_words = ['what', 'who', 'how', 'when', 'which']60 61def get_concepts(text):62    text = text.lower()63    doc = nlp(text)64    concepts = []65    for chunk in doc.noun_chunks:66        if chunk.text not in wh_words:67            concepts.append(chunk.text)68    return concepts69 70def get_passages(text, k=100):71    doc = nlp(text)72    passages = []73    passage_len = 074    passage = ""75    sents = list(doc.sents)76    for i in range(len(sents)):77        sen = sents[i]78        passage_len += len(sen)79        if passage_len >= k:80            passages.append(passage)81            passage = sen.text82            passage_len = len(sen)83            continue84        elif i == (len(sents) - 1):85            passage += " " + sen.text86            passages.append(passage)87            passage = ""88            passage_len = 089            continue90        passage += " " + sen.text91    return passages92 93def get_dicts_for_dpr(concepts, n_results=20, k=100):94    dicts = []95    for concept in concepts:96        wikis = wikipedia.search(concept, results=n_results)97        st.write(f"{concept} No of Wikis: {len(wikis)}")98        for wiki in wikis:99            try:100                html_page = wikipedia.page(title=wiki, auto_suggest=False)101            except DisambiguationError:102                continue103            htmlResults = html_page.content104            passages = get_passages(htmlResults, k=k)105            for passage in passages:106                i_dicts = {}107                i_dicts['text'] = passage108                i_dicts['title'] = wiki109                dicts.append(i_dicts)110    return dicts111 112passage_encoder = TFAutoModel.from_pretrained("nlpconnect/dpr-ctx_encoder_bert_uncased_L-2_H-128_A-2")113query_encoder = TFAutoModel.from_pretrained("nlpconnect/dpr-question_encoder_bert_uncased_L-2_H-128_A-2")114p_tokenizer = AutoTokenizer.from_pretrained("nlpconnect/dpr-ctx_encoder_bert_uncased_L-2_H-128_A-2")115q_tokenizer = AutoTokenizer.from_pretrained("nlpconnect/dpr-question_encoder_bert_uncased_L-2_H-128_A-2")116 117def get_title_text_combined(passage_dicts):118    res = []119    for p in passage_dicts:120        res.append(tuple((p['title'], p['text'])))121    return res122 123def extracted_passage_embeddings(processed_passages, max_length=156):124    passage_inputs = p_tokenizer.batch_encode_plus(125                    processed_passages,126                    add_special_tokens=True,127                    truncation=True,128                    padding="max_length",129                    max_length=max_length,130                    return_token_type_ids=True131                )132    passage_embeddings = passage_encoder.predict([np.array(passage_inputs['input_ids']), np.array(passage_inputs['attention_mask']), 133                                            np.array(passage_inputs['token_type_ids'])], 134                                            batch_size=64, 135                                            verbose=1)136    return passage_embeddings137 138def extracted_query_embeddings(queries, max_length=64):139    query_inputs = q_tokenizer.batch_encode_plus(140        queries,141        add_special_tokens=True,142        truncation=True,143        padding="max_length",144        max_length=max_length,145        return_token_type_ids=True146    )147    148    query_embeddings = query_encoder.predict([np.array(query_inputs['input_ids']),149        np.array(query_inputs['attention_mask']),150        np.array(query_inputs['token_type_ids'])],151        batch_size=1,152        verbose=1)153    return query_embeddings154 155def get_pagetext(page):156    s = str(page).replace("/t","")157    return s158 159def get_wiki_summary(search):160    wiki_wiki = wikipediaapi.Wikipedia('en')161    page = wiki_wiki.page(search)                                   162 163 164def get_wiki_summaryDF(search):165    wiki_wiki = wikipediaapi.Wikipedia('en')166    page = wiki_wiki.page(search)167 168    isExist = page.exists()169    if not isExist:170        return isExist, "Not found", "Not found", "Not found", "Not found"171 172    pageurl = page.fullurl173    pagetitle = page.title174    pagesummary = page.summary[0:60]175    pagetext = get_pagetext(page.text)176 177    backlinks = page.backlinks178    linklist = ""179    for link in backlinks.items():180      pui = link[0]181      linklist += pui + " ,  "182      a=1 183      184    categories = page.categories185    categorylist = ""186    for category in categories.items():187      pui = category[0]188      categorylist += pui + " ,  "189      a=1     190    191    links = page.links192    linklist2 = ""193    for link in links.items():194      pui = link[0]195      linklist2 += pui + " ,  "196      a=1 197      198    sections = page.sections199    200    ex_dic = {201      'Entity' : ["URL","Title","Summary", "Text", "Backlinks", "Links", "Categories"],202      'Value': [pageurl, pagetitle, pagesummary, pagetext, linklist,linklist2, categorylist ]203    }204 205    df = pd.DataFrame(ex_dic)206    207    return df208 209 210def save_message(name, message):211    now = datetime.datetime.now()212    timestamp = now.strftime("%Y-%m-%d %H:%M:%S")213    with open("chat.txt", "a") as f:214        f.write(f"{timestamp} - {name}: {message}\n")215 216def press_release():217    st.markdown("""🎉🎊 Breaking News! 📢📣218Introducing StreamlitWikipediaChat - the ultimate way to chat with Wikipedia and the whole world at the same time! 🌎📚👋219Are you tired of reading boring articles on Wikipedia? Do you want to have some fun while learning new things? Then StreamlitWikipediaChat is just the thing for you! 😃💻220With StreamlitWikipediaChat, you can ask Wikipedia anything you want and get instant responses! Whether you want to know the capital of Madagascar or how to make a delicious chocolate cake, Wikipedia has got you covered. 🍰🌍221But that's not all! You can also chat with other people from around the world who are using StreamlitWikipediaChat at the same time. It's like a virtual classroom where you can learn from and teach others. 🌐👨‍🏫👩‍🏫222And the best part? StreamlitWikipediaChat is super easy to use! All you have to do is type in your question and hit send. That's it! 🤯🙌223So, what are you waiting for? Join the fun and start chatting with Wikipedia and the world today! 😎🎉224StreamlitWikipediaChat - where learning meets fun! 🤓🎈""")225 226 227def main():228    st.title("Streamlit Chat")229 230    name = st.text_input("Enter your name")231    message = st.text_input("Enter a topic to share from Wikipedia")232    if st.button("Submit"):233        234        # wiki235        df = get_wiki_summaryDF(message)236        237        save_message(name, message)238        save_message(name, df)239        240        st.text("Message sent!")241 242    243    st.text("Chat history:")244    with open("chat.txt", "a+") as f:245        f.seek(0)246        chat_history = f.read()247    #st.text(chat_history)248    st.markdown(chat_history)249 250    countdown = st.empty()251    t = 60252    while t:253        mins, secs = divmod(t, 60)254        countdown.text(f"Time remaining: {mins:02d}:{secs:02d}")255        time.sleep(1)256        t -= 1257        if t == 0:258            countdown.text("Time's up!")259            with open("chat.txt", "a+") as f:260                f.seek(0)261                chat_history = f.read()262            #st.text(chat_history)263            st.markdown(chat_history)264 265            press_release()266            267            t = 60268 269if __name__ == "__main__":270    main()271 272