This is a reference to the YouTube tutorial here: https://youtu.be/hSQY4N1u3v0

falcon_pdf_bot.py

import argparse

from pdfminer.high_level import extract_text
from sentence_transformers import SentenceTransformer, CrossEncoder, util

from text_generation import Client

PREPROMPT = "Below are a series of dialogues between various people and an AI assistant. The AI tries to be helpful, polite, honest, sophisticated, emotionally aware, and humble-but-knowledgeable. The assistant is happy to help with almost anything, and will do its best to understand exactly what is needed. It also tries to avoid giving false or misleading information, and it caveats when it isn't entirely sure about the right answer. That said, the assistant is practical and really does its best, and doesn't let caution get too much in the way of being useful.\n"
PROMPT = """"Use the following pieces of context to answer the question at the end.
If you don't know the answer, just say that you don't know, don't try to
make up an answer. Don't make up new terms which are not available in the context.

{context}"""

END_7B = "\n<|prompter|>{query}<|endoftext|><|assistant|>"
END_40B = "\nUser: {query}\nFalcon:"

PARAMETERS = {
    "temperature": 0.9,
    "top_p": 0.95,
    "repetition_penalty": 1.2,
    "top_k": 50,
    "truncate": 1000,
    "max_new_tokens": 1024,
    "seed": 42,
    "stop_sequences": ["<|endoftext|>", "</s>"],
}
CLIENT_7B = Client("http://")  # Fill this part
CLIENT_40B = Client("https://")  # Fill this part


def parse_args():
    parser = argparse.ArgumentParser()
    parser.add_argument("--fname", type=str, required=True)
    parser.add_argument("--top-k", type=int, default=32)
    parser.add_argument("--window-size", type=int, default=128)
    parser.add_argument("--step-size", type=int, default=100)
    return parser.parse_args()


def embed(fname, window_size, step_size):
    text = extract_text(fname)
    text = " ".join(text.split())
    text_tokens = text.split()

    sentences = []
    for i in range(0, len(text_tokens), step_size):
        window = text_tokens[i : i + window_size]
        if len(window) < window_size:
            break
        sentences.append(window)

    paragraphs = [" ".join(s) for s in sentences]
    model = SentenceTransformer("sentence-transformers/all-mpnet-base-v2")
    model.max_seq_length = 512
    cross_encoder = CrossEncoder("cross-encoder/ms-marco-MiniLM-L-6-v2")

    embeddings = model.encode(
        paragraphs,
        show_progress_bar=True,
        convert_to_tensor=True,
    )
    return model, cross_encoder, embeddings, paragraphs


def search(query, model, cross_encoder, embeddings, paragraphs, top_k):
    query_embeddings = model.encode(query, convert_to_tensor=True)
    query_embeddings = query_embeddings.cuda()
    hits = util.semantic_search(
        query_embeddings,
        embeddings,
        top_k=top_k,
    )[0]

    cross_input = [[query, paragraphs[hit["corpus_id"]]] for hit in hits]
    cross_scores = cross_encoder.predict(cross_input)

    for idx in range(len(cross_scores)):
        hits[idx]["cross_score"] = cross_scores[idx]

    results = []
    hits = sorted(hits, key=lambda x: x["cross_score"], reverse=True)
    for hit in hits[:5]:
        results.append(paragraphs[hit["corpus_id"]].replace("\n", " "))
    return results


if __name__ == "__main__":
    args = parse_args()
    model, cross_encoder, embeddings, paragraphs = embed(
        args.fname,
        args.window_size,
        args.step_size,
    )
    print(embeddings.shape)
    while True:
        print("\n")
        query = input("Enter query: ")
        results = search(
            query,
            model,
            cross_encoder,
            embeddings,
            paragraphs,
            top_k=args.top_k,
        )

        query_7b = PREPROMPT + PROMPT.format(context="\n".join(results))
        query_7b += END_7B.format(query=query)

        query_40b = PREPROMPT + PROMPT.format(context="\n".join(results))
        query_40b += END_40B.format(query=query)

        text = ""
        for response in CLIENT_7B.generate_stream(query_7b, **PARAMETERS):
            if not response.token.special:
                text += response.token.text

        print("\n***7b response***")
        print(text)

        text = ""
        for response in CLIENT_40B.generate_stream(query_40b, **PARAMETERS):
            if not response.token.special:
                text += response.token.text

        print("\n***40b response***")
        print(text)

Hi Abhishek,

The sentences.append(window) from line 51 should go before the if condition. Otherwise, sentences will miss the last chunk of the text.

    for i in range(0, len(text_tokens), step_size):
        window = text_tokens[i : i + window_size]
        sentences.append(window)  # Here!
        if len(window) < window_size:
            break

Thanks for sharing it. I discovered this website when looking for a better platform. by Visiting the website link who can do my essay for me. Their essay writers are skilled and knowledgeable, and they always deliver well-researched and original content. I trust them with my assignments and have never been disappointed. Thank you, EduBirdie, for making my college journey smoother!

Hey Abhishek,
Am stuck with "fatal error: cannot execute ‘cc1obj’: execvp: No such file or directory".
Am on Ubuntu. I tried changing permissions of gcc, reinstalling build essentials etc but still getting the same error at compilation.
any idea on how to fix this?

Indeed very nice explanation in the video. I have a request though, I created something like this with vector databases but I was getting very bad responses when the program encounters a pdf document which has both tables and text and the context of the documents in somewhat a dependent on the combination of both. I am not able to figure out how to deal with this kind of problem. Would be a great help if you can suggest something on this line.

@vipulchauhan13 After creating the vector database, how did you do the retrieval using abhishek's way.

Hi Guys,

It's been 2 days am stuck trying to fix "Torch not compiled with CUDA enabled for Embedding". Here is my Development environment on Windows 11, any suggestion would really help me get ahead with this Tutorial from Abhishek.

CUDA Toolkit: 10.1
torch version : 2.3.1
torchaudio version: 2.3.1
transformers version: 4.11.3
python version: 3.8.19
CUDA availability : True

Traceback (most recent call last):
File "d:\Full_Stack_Dev\Machine_Learning\Projects\AI_ML\Telecom_AI_ML\Chatbot_Project\chatbot_pdf.py", line 90, in
results = search(
^^^^^^^
File "d:\Full_Stack_Dev\Machine_Learning\Projects\AI_ML\Telecom_AI_ML\Chatbot_Project\chatbot_pdf.py", line 59, in search
query_embeddings = query_embeddings.cuda()
^^^^^^^^^^^^^^^^^^^^^^^
File "D:\Full_Stack_Dev\Machine_Learning\Projects\AI_ML\Telecom_AI_ML\Chatbot_Project\chatbot\Lib\site-packages\torch\cuda_init_.py", line 239, in _lazy_init
raise AssertionError("Torch not compiled with CUDA enabled")
AssertionError: Torch not compiled with CUDA enabled

Cheers
Amit K