import urllib.request import fitz import re import numpy as np import tensorflow_hub as hub import openai import gradio as gr import os from sklearn.neighbors import NearestNeighbors def download_pdf(url, output_path): urllib.request.urlretrieve(url, output_path) def preprocess(text): text = text.replace('\n', ' ') text = re.sub('\s+', ' ', text) return text def pdf_to_text(path, start_page=1, end_page=None): doc = fitz.open(path) total_pages = doc.page_count if end_page is None: end_page = total_pages text_list = [] for i in range(start_page-1, end_page): text = doc.load_page(i).get_text("text") text = preprocess(text) text_list.append(text) doc.close() return text_list def text_to_chunks(texts, word_length=150, start_page=1): text_toks = [t.split(' ') for t in texts] page_nums = [] chunks = [] for idx, words in enumerate(text_toks): for i in range(0, len(words), word_length): chunk = words[i:i+word_length] if (i+word_length) > len(words) and (len(chunk) < word_length) and ( len(text_toks) != (idx+1)): text_toks[idx+1] = chunk + text_toks[idx+1] continue chunk = ' '.join(chunk).strip() chunk = f'[Page no. {idx+start_page}]' + ' ' + '"' + chunk + '"' chunks.append(chunk) return chunks class SemanticSearch: def __init__(self): self.use = hub.load('https://tfhub.dev/google/universal-sentence-encoder/4') self.fitted = False def fit(self, data, batch=1000, n_neighbors=5): self.data = data self.embeddings = self.get_text_embedding(data, batch=batch) n_neighbors = min(n_neighbors, len(self.embeddings)) self.nn = NearestNeighbors(n_neighbors=n_neighbors) self.nn.fit(self.embeddings) self.fitted = True def __call__(self, text, return_data=True): inp_emb = self.use([text]) neighbors = self.nn.kneighbors(inp_emb, return_distance=False)[0] if return_data: return [self.data[i] for i in neighbors] else: return neighbors def get_text_embedding(self, texts, batch=1000): embeddings = [] for i in range(0, len(texts), batch): text_batch = texts[i:(i+batch)] emb_batch = self.use(text_batch) embeddings.append(emb_batch) embeddings = np.vstack(embeddings) return embeddings def load_recommender(path, start_page=1): global recommender texts = pdf_to_text(path, start_page=start_page) chunks = text_to_chunks(texts, start_page=start_page) recommender.fit(chunks) return 'Corpus Loaded.' def generate_text(openAI_key, prompt, model="gpt-3.5-turbo"): openai.api_key = openAI_key if model == "text-davinci-003": completions = openai.Completion.create( engine=model, prompt=prompt, max_tokens=512, n=1, stop=None, temperature=0.7, ) message = completions.choices[0].text else: message = openai.ChatCompletion.create( model=model, messages=[ {"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": prompt} ] ).choices[0].message['content'] return message def generate_answer(question, openAI_key,model): topn_chunks = recommender(question) prompt = "" prompt += 'search results:\n\n' for c in topn_chunks: prompt += c + '\n\n' prompt += "Instructions: Compose a comprehensive reply to the query using the search results given. "\ "Cite each reference using [ Page Number] notation (every result has this number at the beginning). "\ "Citation should be done at the end of each sentence. If the search results mention multiple subjects "\ "with the same name, create separate answers for each. Only include information found in the results and "\ "don't add any additional information. Make sure the answer is correct and don't output false content. "\ "If the text does not relate to the query, simply state 'Found Nothing'. Ignore outlier "\ "search results which has nothing to do with the question. Only answer what is asked. The "\ "answer should be short and concise. \n\nQuery: {question}\nAnswer: " prompt += f"Query: {question}\nAnswer:" answer = generate_text(openAI_key, prompt, model) return answer def question_answer(url, file, question, openAI_key, model): try: if openAI_key.strip()=='': return '[ERROR]: Please enter you Open AI Key. Get your key here : https://platform.openai.com/account/api-keys' if url.strip() == '' and file == None: return '[ERROR]: Both URL and PDF is empty. Provide at least one.' if url.strip() != '' and file != None: return '[ERROR]: Both URL and PDF is provided. Please provide only one (either URL or PDF).' if model is None or model =='': return '[ERROR]: You have not selected any model. Please choose an LLM model.' if url.strip() != '': glob_url = url download_pdf(glob_url, 'corpus.pdf') load_recommender('corpus.pdf') else: old_file_name = file.name file_name = file.name file_name = file_name[:-12] + file_name[-4:] os.rename(old_file_name, file_name) load_recommender(file_name) if question.strip() == '': return '[ERROR]: Question field is empty' if model == "text-davinci-003": return generate_answer_text_davinci_003(question, openAI_key) else: return generate_answer(question, openAI_key, model) except openai.error.InvalidRequestError as e: return f'[ERROR]: Either you do not have access to GPT4 or you have exhausted your quota!' def generate_text_text_davinci_003(openAI_key,prompt, engine="text-davinci-003"): openai.api_key = openAI_key completions = openai.Completion.create( engine=engine, prompt=prompt, max_tokens=512, n=1, stop=None, temperature=0.7, ) message = completions.choices[0].text return message def generate_answer_text_davinci_003(question,openAI_key): topn_chunks = recommender(question) prompt = "" prompt += 'search results:\n\n' for c in topn_chunks: prompt += c + '\n\n' prompt += "Instructions: Compose a comprehensive reply to the query using the search results given. "\ "Cite each reference using [ Page Number] notation (every result has this number at the beginning). "\ "Citation should be done at the end of each sentence. If the search results mention multiple subjects "\ "with the same name, create separate answers for each. Only include information found in the results and "\ "don't add any additional information. Make sure the answer is correct and don't output false content. "\ "If the text does not relate to the query, simply state 'Found Nothing'. Ignore outlier "\ "search results which has nothing to do with the question. Only answer what is asked. The "\ "answer should be short and concise. \n\nQuery: {question}\nAnswer: " prompt += f"Query: {question}\nAnswer:" answer = generate_text_text_davinci_003(openAI_key, prompt,"text-davinci-003") return answer recommender = SemanticSearch() title = 'PDF GPT Turbo' description = """ PDF GPT Turbo allows you to chat with your PDF file using Universal Sentence Encoder and Open AI. It gives hallucination free response than other tools as the embeddings are better than OpenAI. The returned response can even cite the page number in square brackets([]) where the information is located, adding credibility to the responses and helping to locate pertinent information quickly.""" with gr.Blocks() as demo: gr.Markdown(f'

{title}

') gr.Markdown(description) with gr.Row(): with gr.Group(): gr.Markdown(f'

Get your Open AI API key here

') #openAI_key=gr.Textbox(label='Enter your OpenAI API key here') openAI_key = gr.Textbox(label='Enter your OpenAI API key here', password=True) url = gr.Textbox(label='Enter PDF URL here') gr.Markdown("

OR

") file = gr.File(label='Upload your PDF/ Research Paper / Book here', file_types=['.pdf']) question = gr.Textbox(label='Enter your question here') model = gr.Radio(['gpt-3.5-turbo', 'gpt-3.5-turbo-16k', 'gpt-3.5-turbo-0613', 'gpt-3.5-turbo-16k-0613', 'text-davinci-003','gpt-4','gpt-4-32k'], label='Select Model', default='gpt-3.5-turbo') #model = gr.Dropdown(choices=['gpt-3.5-turbo', 'gpt-3.5-turbo-16k', 'gpt-3.5-turbo-0613', 'gpt-3.5-turbo-16k-0613', 'text-davinci-003'], label='Select Large Language Model', default='gpt-3.5-turbo') btn = gr.Button(value='Submit') btn.style(full_width=True) with gr.Group(): answer = gr.Textbox(label='The answer to your question is :') #btn.click(question_answer, inputs=[url, file, question,openAI_key], outputs=[answer]) btn.click(question_answer, inputs=[url, file, question, openAI_key, model], outputs=[answer]) #openai.api_key = os.getenv('Your_Key_Here') demo.launch()