Academic_Papers_Writing.py

# -*- coding: utf-8 -*-
"""Academic_Papers_Writing_translat.ipynb

Automatically generated by Colaboratory.

Original file is located at
    https://colab.research.google.com/github/So-AI-love/chatgpt-prompts-for-academic-writing/blob/main/Auto_Making/Academic_Papers_Writing_translat.ipynb

## Aromatically wrrie apaper by the ChatGPT and this github prompt instruction:

https://github.com/ahmetbersoz/chatgpt-prompts-for-academic-writing
"""

!pip install openai
!pip install backoff
!pip install docx2pdf
!pip install python-docx
!pip install django

# SETUP COLAB for run Streamlit
!npm install localtunnel

!pip install -q dl-translate
!curl ipv4.icanhazip.com

!sudo apt-get update
!sudo apt-get install libreoffice
#!sudo apt-get install abiword
!sudo apt install libreoffice-writer

# Define main variables
TOPIC = "{TOPIC}"
RESEARCH_DOMAIN = "{RESEARCH_DOMAIN}"
PARAGRAPH = "{PARAGRAPH}"
PARAGRAPHS = "{PARAGRAPHS}"
TOPIC_SENTENCE = "{TOPIC_SENTENCE}"
LANGUAGE = "{LANGUAGE}"
ABSTRACT_PARAGRAPH = "{ABSTRACT_PARAGRAPH}"
BIBLIOGRAPHY = "{BIBLIOGRAPHY}"
THEORY1 = "{THEORY1}"
THEORY2 = "{THEORY2}"
RESEARCH_QUESTIONS = "{RESEARCH_QUESTIONS}"
ACTION = "{ACTION}"
RESULT_PARAGRAPHS = "{RESULT_PARAGRAPHS}"
DATE = "{DATE}"
NUMBER_OF_DAYS_MONTHS_YEARS = "{NUMBER_OF_DAYS_MONTHS_YEARS}"

# List of prompts for generating academic papers
prompts = [

  # Improving Language
  f"Write a counterargument to the following claim: '{PARAGRAPH}'",
  f"Rewrite this in an academic voice: '{PARAGRAPH}'",
  f"Expand these notes: '{PARAGRAPH}'",
  f"Provide me a list of words and phrases which were repeatedly / more than 3 times used: '{PARAGRAPHS}'",
  f"Provide me a list of synonyms for '{PARAGRAPH}' and evaluate them in the context of '{PARAGRAPH}'",
  f"Act as a language expert, proofread my paper on '{TOPIC_SENTENCE}' while putting a focus on grammar and punctuation.",
  f"In the context of '{RESEARCH_DOMAIN}' translate '{PARAGRAPH}' into the '{LANGUAGE}' language.",

   # Brainstorming
   f"Find a research topic for a PhD in the area of '{TOPIC}'",
   f"Write a detailed proposal on the following research topic. Make sure it is free from plagiarism. '{PARAGRAPH}'",
   f"Identify gaps in the literature on '{TOPIC_SENTENCE}'",
   f"Generate 10 academic research questions about '{PARAGRAPHS}'",
   f"Generate a list of research hypotheses related to '{TOPIC_SENTENCE}'",
   f"Identify potential areas for future research in the context of this '{TOPIC_SENTENCE}'",
   f"Suggest novel applications of '{TOPIC_SENTENCE}' within '{RESEARCH_DOMAIN}'",

   # Title/Topic Sentence
   f"Suggest 5 titles for the following abstract: '{ABSTRACT_PARAGRAPH}'",
   f"Write a topic sentence for this paragraph: '{PARAGRAPH}'",

   # Keywords
   f"Provide 5 keywords for this: '{PARAGRAPHS}'",

   # Abstract
   f"Generate an abstract for a scientific paper based on this information for: '{PARAGRAPHS}'",

   # Outline
   f"Generate an outline for '{TOPIC_SENTENCE}'",
   f"I want to write a journal article about '{TOPIC_SENTENCE}'. Give me an outline for the article that I can use as a starting point.",

   # Introduction
   f"Come up with an introduction for the following research topic: '{TOPIC_SENTENCE}'",

   # Literature Review
   f"Conduct a literature review on '{TOPIC_SENTENCE}' and provide review paper references",
   f"Provide me with references and links to papers in '{PARAGRAPH}'",
   f"Summarize the scholarly literature including in-text citations on '{PARAGRAPHS}'",
   f"Write this in standard Harvard referencing '{PARAGRAPH}'",
   f"Convert this '{BIBLIOGRAPHY}' from MLA to APA style.",
   f"Compare and contrast '{THEORY1}' and '{THEORY2}' in the context of '{RESEARCH_DOMAIN}'",

   # Methodology
   f"Create objectives and methodology for '{TOPIC_SENTENCE}'",
   f"Write a detailed methodology for the topic: '{TOPIC_SENTENCE}'",
   f"Analyze the strengths and weaknesses of this methodology: '{PARAGRAPHS}'",
   f"Write objectives for this study: '{TOPIC_SENTENCE}'",
   f"What are the limitations of using '{TOPIC_SENTENCE}' in '{RESEARCH_DOMAIN}'?",
   f"Create a recipe for the methods used in this '{PARAGRAPHS}'",
   f"Suggest interdisciplinary approaches to '{TOPIC_SENTENCE}'",
   f"Explain how qualitative/quantitative research methods can be used to address '{RESEARCH_QUESTIONS}'",
   f"Recommend best practices for data collection and analysis in '{TOPIC_SENTENCE}'",

   # Experiments
   f"Design an experiment that '{ACTION}'",

   # Results
   f"Write a result section for the following paragraphs. Please write this in the third person. '{PARAGRAPHS}'",

   # Discussion
   f"Discuss this results: '{RESULT_PARAGRAPHS}'",

   # Conclusion
   f"Generate a conclusion for this: '{PARAGRAPHS}'",
   f"Give recommendations and conclusion for: '{PARAGRAPHS}'",

   # Future Works
   f"Can you suggest 3 directions for future research on this topic: '{PARAGRAPH}'",

   # Plan/Presentation
   f"Develop a research plan for: '{TOPIC_SENTENCE}'",
   f"Write a schedule for completion in '{TOPIC_SENTENCE}' in NUMBER OF DAYS MONTHS YEARS which is '{NUMBER_OF_DAYS_MONTHS_YEARS}'",
   f"The deadline for the submission of the first draft is '{DATE}'. Give me a week-by-week breakdown so I can plan my writing better.",
   f"Write a sensational press release for this research: '{PARAGRAPHS}'",
   f"Make this more persuasive: '{PARAGRAPH}'",
   f"Write 3 tweets about this research? '{PARAGRAPHS}'",
]

TOPIC = f"Understanding the Cycle of Domestic Violence against iranian women: The Influence of changes in the type of violence that continues in the aftermath of violence"

#TOPIC = f" "

"""#Funstions for saving as PDF and DOCX :
👇🌱
"""

!sudo apt-get update
!sudo apt-get install libreoffice
#!sudo apt-get install abiword
!sudo apt install libreoffice-writer

import subprocess

def convert_docx_to_pdf0(docx_path, pdf_path):
  subprocess.call(['abiword', '--to=pdf', docx_path, '--to-dir', pdf_path])

def convert_docx_to_pdf(docx_path, pdf_path):
 command = ['libreoffice', '--headless', '--convert-to', 'pdf', '--outdir', pdf_path, docx_path]
 subprocess.call(command)

import os
import subprocess
from docx import Document



def save_academic_paper(topic, prompt_my):
 # Define the path to the DOCX file
 docx_path = f"/content/{topic}.docx"

 # Check if the DOCX file exists
 if os.path.isfile(docx_path):
     # If the DOCX file exists, open it
     doc = Document(docx_path)
 else:
     # If the DOCX file doesn't exist, create a new Document object
     doc = Document()

 # Add the generated text to the document
 #print ("____&&&&&&&&&&\n",prompt_my)
 doc.add_paragraph(prompt_my)

 # Save the document
 doc.save(docx_path)

 # Convert the DOCX file to a PDF
 convert_docx_to_pdf(docx_path, "/content/output/")

from google.colab import drive
import os
import subprocess
from docx import Document
from django.utils.text import slugify

# Mount Google Drive
drive.mount('/content/drive')

def save_academic_paper(topic, prompt_my):
  # Define the path to the folder in Google Drive
  folder_path = r"/content/drive/My Drive/ChatGPT_Paper_wrting/"

  Pdf_Dir= r"/content/drive/My Drive/ChatGPT_Paper_wrting/"
  docx_path= f"{folder_path}{topic}.docx"
  # Check if the folder exists
  if not os.path.exists(folder_path):
  # If the folder doesn't exist, create it
    os.mkdir(folder_path)


  # Replace spaces in the topic with underscores
  #topic = topic[:12].replace(" ", "_")
  topic = slugify(topic)

  # Define the path to the DOCX file in the folder


  # Check if the DOCX file exists
  if os.path.isfile(docx_path):
    # If the DOCX file exists, open it

    #docx_path = f"{folder_path}{topic}.docx"#+str(random.randint(0,9))+".docx"
    doc = Document(docx_path)
  else:

    #docx_path = f"{folder_path}{topic}.docx"
    #print(docx_path)
     # If the DOCX file doesn't exist, create a new Document object
    doc = Document()

  # Add the generated text to the document
  doc.add_paragraph(prompt_my)

  # Save the document
  doc.save(docx_path)

  # Convert the DOCX file to a PDF
  convert_docx_to_pdf(docx_path,Pdf_Dir)

#save_academic_paper("Persian_"+'topic','\n**'+'choice_translated_prompt'+'**\n'+"choice_translated")
topic = TOPIC[:20].replace(" ", "_")
topic = slugify(TOPIC[:-5])
docx_path = f"{topic}.docx"

print(docx_path)

#!pip install googletrans==4.0.0-rc1

!pip install httpcore==0.15.0 httpx pymongo googletrans
!pip install httpx==0.24.1

"""#I have tried to update the prompt by ChatGPT itself 👇👇"""

def extract_main_variables(prompts, variable):
   main_variables = {
       'TOPIC': None,
       'RESEARCH_DOMAIN': None,
       'PARAGRAPH': None,
       'PARAGRAPHS': None,
       'TOPIC_SENTENCE': None,
       'LANGUAGE': None,
       'ABSTRACT_PARAGRAPH': None,
       'BIBLIOGRAPHY': None,
       'THEORY1': None,
       'THEORY2': None,
       'RESEARCH_QUESTIONS': None,
       'ACTION': None,
       'RESULT_PARAGRAPHS': None,
       'DATE': None,
       'NUMBER_OF_DAYS_MONTHS_YEARS': None
   }

   for prompt in prompts:
       variables = re.findall(r'\{(\w+)\}', prompt)
       for var in variables:
           if var == 'TOPIC':
               main_variables[var] = variable
           else:
               main_variables[var] = None

   return main_variables

import re

def generate_prompt_from_response(previous_response, main_variables):
   # Customize this logic based on your needs
   # For simplicity, let's use the last 50 characters of the response as the new prompt
   new_prompt = previous_response[-50:]

   # Replace the main variables in the new prompt
   for var, value in main_variables.items():
       if value is not None:
           new_prompt = new_prompt.replace(value, f"{{{var}}}")

   return new_prompt

# Example usage
#TOPIC =f"strategies for increase the compassion in iranian Women movment and ist relationship with the fatigue of compassion in light triad personality"  #f" the dark triad in psychology and it's relation with {goal}"
#TOPIC = f"Understanding the Cycle of Domestic Violence against iranian women: The Influence of Economic Power"

prompts_1 = [
f"Find a research topic for a PhD in the area of '{TOPIC}'",

]
#prompts_2 = [
#   "I need to find information on renewable energy for a research paper that will be 1,500 words and must include eight sources.",
#   "The research paper should cover the latest advancements in renewable energy technology."
#]

main_variables = extract_main_variables(prompts,TOPIC)
print(main_variables)

previous_response = "Previous GPT response"
new_prompt = generate_prompt_from_response(previous_response, main_variables)
print(new_prompt)
print ( prompts)

"""#Updating the prompt by ChatGPT:👇👇"""

def generate_prompt_update_0(prompt, previous_content, main_variables):
   model_engine = "text-davinci-003"
   max_tokens = 2048

   # Replace the TOPIC variable in the prompt
   for var, value in main_variables.items():
       if var == 'TOPIC' and value is not None:
           prompt = prompt.replace(f"{{{var}}}", value)

   # Construct the instruction for updating the prompt
   instruction = f"Given the previous content:\n\n{previous_content}\n\nUpdate the prompt: {prompt}"

   # Create the completion with the instruction
   completion = client.completions.create(
       model=model_engine,
       prompt=instruction,
       max_tokens=max_tokens,
       temperature=0.3,
       n=1,
       frequency_penalty=0,
   )

   # Extract and return the updated prompt from the response
   updated_prompt = completion.choices[0].text.strip()
   return updated_prompt

from openai import RateLimitError
from backoff import on_exception, expo

@on_exception(expo, RateLimitError, max_tries=20, max_time=600)
def generate_prompt_update_a1(prompt_my):
  model_engine = "text-davinci-003"
  max_tokens = 2048

  completion = client.completions.create(
      model="gpt-3.5-turbo-instruct",
      prompt=prompt_my,
      max_tokens=2048,
      temperature=0.3,
      n=1,
      frequency_penalty=0,
  )
  return completion

import random
import time
import openai
from openai import RateLimitError
import os
from openai import OpenAI

#openai.api_key = "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"# "your_openai_api_key"

client = OpenAI(
   # defaults to os.environ.get("OPENAI_API_KEY")
   api_key = "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",
)
def retry_with_exponential_backoff(
   func,
   initial_delay: float = 1,
   exponential_base: float = 2,
   jitter: bool = True,
   max_retries: int = 10,
   errors: tuple = (RateLimitError,),
):
   """Retry a function with exponential backoff."""
   def wrapper(*args, **kwargs):
       # Initialize variables
       num_retries = 0
       delay = initial_delay
       # Loop until a successful response or max_retries is hit or an exception is raised
       while True:
           try:
               return func(*args, **kwargs)
           # Retry on specified errors
           except errors as e:
               # Increment retries
               num_retries += 1
               # Check if max retries has been reached
               if num_retries > max_retries:
                  raise Exception(
                      f"Maximum number of retries ({max_retries}) exceeded."
                  )
               # Increment the delay
               delay *= exponential_base * (1 + jitter * random.random())
               # Sleep for the delay
               time.sleep(delay)
           # Raise exceptions for any errors not specified
           except Exception as e:
               raise e
   return wrapper
from openai import RateLimitError
from backoff import on_exception, expo

@on_exception(expo, RateLimitError, max_tries=20, max_time=600)

@retry_with_exponential_backoff
def generate_prompt_update(prompt_my):
 model_engine = "text-davinci-003"
 max_tokens = 2048

 completion = client.completions.create(
     model="gpt-3.5-turbo-instruct",
     prompt=prompt_my,
     max_tokens=1024,
     temperature=0.1,
     n=1,
     frequency_penalty=0,
 )
 return completion


@retry_with_exponential_backoff
def generate_academic_paper(prompt_my):
 model_engine = "text-davinci-003"
 max_tokens = 2048

 completion = client.completion.create(
     model="gpt-3.5-turbo-instruct",
     prompt=prompt_my,
     max_tokens=2048,
     temperature=0.3,
     n=1,
     frequency_penalty=0,
 )
 return completion

def generate_content(prompts, variables, TOPIC, perviuse_content=""):
  # Generate content for each variable
  variable_contents = {}

  for prompt in prompts:
      for var in variables:
          if var in prompt:
              print(f"\n---prompt is ----\n{prompt}")
              prompt_my = f"Generate content for the variable '{var}' in the context of the topic '{TOPIC}'. Please consider the result must be in less than 10 words. The prompt is '{prompt}'. Also the more information for understanding better content is '{perviuse_content}'"
              content = generate_prompt_update_a1(prompt_my)
              for choice in content.choices:
                print("\n Result is :"+choice.text)
                variable_contents[var] = choice.text
              print(f"variable_contents[{var}] is: {variable_contents[var]}")

  # Replace the variables in the prompts with the generated content
  prompt_new = []
  for prompt in prompts:
      for var, content in variable_contents.items():
          prompt = prompt.replace(f"{{{var}}}", variable_contents[var])
      prompt_new.append(prompt)
      print("\n --- Updated prompt is :\n"+str(prompt_new))
  return prompt_new

# Define main variables
#TOPIC = "strategies for increase the compassion in iranian Women movment and ist relationship with the fatigue of compassion in light triad personality"

# List of variables
variables = ["RESEARCH_DOMAIN", "PARAGRAPH", "PARAGRAPHS", "TOPIC_SENTENCE", "LANGUAGE", "ABSTRACT_PARAGRAPH", "BIBLIOGRAPHY", "THEORY1", "THEORY2", "RESEARCH_QUESTIONS", "ACTION", "RESULT_PARAGRAPHS", "DATE", "NUMBER_OF_DAYS_MONTHS_YEARS"]

# List of prompts
#prompts = ["prompt1", "prompt2", "prompt3"] # replace with your actual prompts
#perviuse_content= "test"
# Call the function and store the returned prompts
#updated_prompts = generate_content(prompts, variables, TOPIC, perviuse_content)
#print("Updated Prompts:", updated_prompts)

# Import the necessary libraries and set up the API key
#!pip install openai
import json
#TOPIC =f"strategies for increase the compassion in iranian Women movment and ist relationship with the fatigue of compassion in light triad personality" #f" the dark triad in psychology and it's relation with {goal}"
#TOPIC = f"Understanding the Cycle of Domestic Violence against iranian women: The Influence of Economic Power"


#topic = f"The Importance of Focusing on Waste Collection in a Waste-Filled World"
import openai
import os
from openai import OpenAI

#openai.api_key = "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"# "your_openai_api_key"

#client = OpenAI(
   # defaults to os.environ.get("OPENAI_API_KEY")
   #api_key = "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",
#)
#openai.api_key = os.getenv('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')
# Define the prompts
prompts3 = [
f"Find a research topic for a PhD in the area of '{topic}'",
f"Write a detailed proposal on the following research '{topic}'. Make Sure it is free from plagiarism. ",
f"Identify gaps in the literature on '{topic}'",
"Generate 10 academic research questions about Perviuse action",
f"Generate a list of research hypotheses related to '{topic}'"
]

from googletrans import Translator

# Create a Translator object
translator = Translator()

from openai import RateLimitError
from backoff import on_exception, expo

@on_exception(expo, RateLimitError, max_tries=20, max_time=600)
def generate_academic_paper_a0(prompt_my):
  model_engine = "text-davinci-003"
  max_tokens = 2048

  completion = client.completions.create(
      model="gpt-3.5-turbo-instruct",
      prompt=prompt_my,
      max_tokens=2048,
      temperature=0.3,
      n=1,
      frequency_penalty=0,
  )
  return completion

import time,random
from openai import OpenAI

#openai.api_key = "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"# "your_openai_api_key"

client = OpenAI(
   # defaults to os.environ.get("OPENAI_API_KEY")
   api_key = "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",
)

def generate_papers(prompts, perviuse_content, perviuse_try_numner):
  choice_text_all=[]

  for i in range(perviuse_try_numner, len(prompts), 20):
     # Slice the prompts list to get the next 20 prompts
     print("I is ",i," Len of prompt Is:", len(prompts))

     batch = prompts[i:i+20]
     print("batch is ",batch)

     for prompt in batch:

         # Print the prompt
         print("prompt is ",list({prompt}))

         updated_prompts = generate_content(list({prompt}), variables, TOPIC, perviuse_content)
         print("Updated Prompts:", updated_prompts)

         time.sleep(random.randint(22, 40))
         response = generate_academic_paper_a0(updated_prompts)
         print("\nGenerated Academic Paper:")
         print("========================\n")
         for choice in response.choices:
           print(choice.text)
           choice_in_loop = choice.text
           choice_text_all.append(choice.text)


         #save_academic_paper(topic,'\n--------**\n'+updated_prompts+'/n-------**\n'+choice.text)
         save_academic_paper(TOPIC[:10000],'\n--------**\n'+''.join(updated_prompts)+'/n-------**\n'+choice.text)
         if hasattr(choice, 'choices'):
             extract_text(choice)

         perviuse_content = choice_in_loop
         print("\n end of loop")
         print("========================\n")

         time.sleep(random.randint(22, 40))
         perviuse_try_numner = perviuse_try_numner+1
  return choice_text_all,perviuse_try_numner

#TOPIC =f"strategies for increase the compassion in iranian Women movment and ist relationship with the fatigue of compassion in light triad personality" #f" the dark triad in psychology and it's relation with {goal}"
#TOPIC = f"Understanding the Cycle of Domestic Violence against iranian women: The Influence of Economic Power"

# Generate academic papers for the given prompts

def main_generate_papers(prompts,perviuse_content,perviuse_try_numner):

  if not perviuse_try_numner:
     perviuse_try_numner=0
     perviuse_content=['fist step']

  else if perviuse_try_numner= Len (prompts):
     perviuse_try_numner=0

  generate_papers(prompts, perviuse_content,perviuse_try_numner)

return perviuse_content,perviuse_try_numner

main_generate_papers(prompts,perviuse_content,perviuse_try_numner)

"""#Only update the Prompt by the help of ChatGPT:👇👇"""

# Define main variables
#TOPIC = "strategies for increase the compassion in iranian Women movment and ist relationship with the fatigue of compassion in light triad personality"

# List of variables
variables = ["RESEARCH_DOMAIN", "PARAGRAPH", "PARAGRAPHS", "TOPIC_SENTENCE", "LANGUAGE", "ABSTRACT_PARAGRAPH", "BIBLIOGRAPHY", "THEORY1", "THEORY2", "RESEARCH_QUESTIONS", "ACTION", "RESULT_PARAGRAPHS", "DATE", "NUMBER_OF_DAYS_MONTHS_YEARS"]

# Generate content for each variable
variable_contents = {}
for prompt in prompts:
   for var in variables:
       if var in prompt:
           prompt = f"Generate content for the variable '{var}' based on the topic '{TOPIC}'. For this prompt '{prompt}'"
           content= generate_academic_paper(prompt)
           for choice in content.choices:
               print(choice.text)
               variable_contents[var] = choice.text
           print("variable_contents[var] is: ",variable_contents[var])

# Replace the variables in the prompts with the generated content
for prompt in prompts:
   for var, content in variable_contents.items():
       prompt = prompt.replace(f"{{{var}}}", content)
print("Updated Prompt:", prompt)

def generate_prompt_update_a3(prompt, previous_content):
    model_engine = "text-davinci-003"
    max_tokens = 2048

    # Construct the instruction for updating the prompt
    instruction = f"Given the previous content:\n\n{previous_content}\n\nUpdate the prompt: {prompt}"

    # Create the completion with the instruction
    completion = client.completions.create(
        model=model_engine,
        prompt=instruction,
        max_tokens=max_tokens,
        temperature=0.3,
        n=1,
        frequency_penalty=0,
    )

    # Extract and return the updated prompt from the response
    updated_prompt = completion.choices[0].text.strip()
    return updated_prompt
# Inside your loop
for prompt in prompts:
    print("Original Prompt:", prompt)
    response = generate_academic_paper_a2(prompt)
    previous_content = response.choices[0].text.strip()
    updated_prompt = generate_prompt_update_2(prompt, previous_content)
    print("Updated Prompt:", updated_prompt)

    # Continue with the rest of your processing
    # ...

"""# with no translation for become faster answer:👇👇🙏"""

# Import the necessary libraries and set up the API key
#!pip install openai
import json
#TOPIC =f"strategies for increase the compassion in iranian Women movment and ist relationship with the fatigue of compassion in light triad personality"  #f" the dark triad in psychology and it's relation with {goal}"
#TOPIC = f"Understanding the Cycle of Domestic Violence against iranian women: The Influence of Economic Power"
#topic = f"The Importance of Focusing on Waste Collection in a Waste-Filled World"
import openai
import os
from openai import OpenAI

#openai.api_key = "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"# "your_openai_api_key"

client = OpenAI(
    # defaults to os.environ.get("OPENAI_API_KEY")
    api_key = "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",
)
#openai.api_key = os.getenv('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')
# Define the prompts
prompts3 = [
f"Find a research topic for a PhD in the area of '{topic}'",
f"Write a detailed proposal on the following research '{topic}'. Make Sure it is free from plagiarism. ",
f"Identify gaps in the literature on '{topic}'",
"Generate 10 academic research questions about Perviuse action",
f"Generate a list of research hypotheses related to '{topic}'"
]

from googletrans import Translator

# Create a Translator object
translator = Translator()


# Function to generate an academic paper

from openai import RateLimitError
from backoff import on_exception, expo

@on_exception(expo, RateLimitError, max_tries=20, max_time=600)
def generate_academic_paper_a2(prompt_my):
   model_engine = "text-davinci-003"
   max_tokens = 2048

   completion = client.completions.create(
       model="gpt-3.5-turbo-instruct",
       prompt=prompt_my,
       max_tokens=2048,
       temperature=0.3,
       n=1,
       frequency_penalty=0,
   )
   return completion

choice_text_all=[]

import time
# Function to generate academic papers for given prompts
# Function to generate academic papers for given prompts
def generate_papers(prompts):
   for i in range(0, len(prompts), 20):
       # Slice the prompts list to get the next 20 prompts
       print("I is ",i," Len of prompt Is:", len(prompts))

       batch = prompts[i:i+20]
       print("batch is ",batch)
   #for j in range ( 0, Len(batch)):
       # Generate papers for the next 20 prompts
       for prompt in batch:
      #   for prompt in prompt1:
           # Print the prompt
           print("prompt is ", prompt)
           response = generate_academic_paper_a2(prompt)
           print("\nGenerated Academic Paper:")
           print("========================\n")
           #print(response)
           for choice in response.choices:
             print(choice.text)
             choice_text_all.append(choice.text)

             # Translate the generated text to Persian
         #  choice_translated = translate_to_persian(choice.text) #translator.translate(choice.text, dest='fa')
         # choice_translated_prompt = translate_to_persian(prompt)#translator.translate(prompt, dest='fa')

         #    print("\n-----\n Translated is ",choice_translated)
         #    save_academic_paper("Persian_"+topic,'\n**'+choice_translated_prompt+'**\n'+choice_translated)
             save_academic_paper(topic,'\n--------**\n'+prompt+'/n-------**\n'+choice.text)
             # Recursively call the function for the nested Completion objects
             if hasattr(choice, 'choices'):
               extract_text(choice)

           #generated_text = response.choices[0].text
           # Print the generated text
           #print(generated_text)
           print("\n end of loop")
           print("========================\n")

           #print("loop")
       # Wait for a short period of time before sending the next batch of prompts
           time.sleep(40)
   return choice_text_all#,choice.translated

# Generate academic papers for the given prompts
generate_papers(prompts)

# SETUP COLAB for run Streamlit
!npm install localtunnel
!curl ipv4.icanhazip.com

"""# DL Translate
A deep learning-based translation library built on Huggingface transformers and Facebook's mBART-Large

https://colab.research.google.com/github/xhluca/dl-translate/blob/main/demos/colab_demo.ipynb#scrollTo=qdefSjR_YIiG
"""

pip install -q dl-translate

import dl_translate as dlt
mt = dlt.TranslationModel('mbart50')

import dl_translate as dlt

def translate_to_persian(text):
  # Initialize the translation model
  #mt = dlt.TranslationModel('mbart50')

  # Translate the text
  translated = mt.translate(text, source=dlt.lang.ENGLISH, target=dlt.lang.PERSIAN)

  return translated

# Import the necessary libraries and set up the API key
#!pip install openai
import json
#TOPIC =f"strategies for increase the compassion in iranian Women movment and ist relationship with the fatigue of compassion in light triad personality"  #f" the dark triad in psychology and it's relation with {goal}"
#topic = f"The Importance of Focusing on Waste Collection in a Waste-Filled World"
import openai
import os
from openai import OpenAI

#openai.api_key = "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"# "your_openai_api_key"

client = OpenAI(
    # defaults to os.environ.get("OPENAI_API_KEY")
    api_key = "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",
)
#openai.api_key = os.getenv('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')
# Define the prompts
prompts3 = [
f"Find a research topic for a PhD in the area of '{topic}'",
f"Write a detailed proposal on the following research '{topic}'. Make Sure it is free from plagiarism. ",
f"Identify gaps in the literature on '{topic}'",
"Generate 10 academic research questions about Perviuse action",
f"Generate a list of research hypotheses related to '{topic}'"
]

from googletrans import Translator

# Create a Translator object
translator = Translator()


# Function to generate an academic paper

from openai import RateLimitError
from backoff import on_exception, expo

@on_exception(expo, RateLimitError, max_tries=10, max_time=600)
def generate_academic_paper_5(prompt_my):
   model_engine = "text-davinci-003"
   max_tokens = 2048

   completion = client.completions.create(
       model="gpt-3.5-turbo-instruct",
       prompt=prompt_my,
       max_tokens=2048,
       temperature=0.3,
       n=1,
       frequency_penalty=0,
   )
   return completion

choice_text_all=[]

import time
# Function to generate academic papers for given prompts
# Function to generate academic papers for given prompts
def generate_papers(prompts):
   for i in range(0, len(prompts), 20):
       # Slice the prompts list to get the next 20 prompts
       print("I is ",i," Len of prompt Is:", len(prompts))
       batch = prompts[i:i+10]
       print("batch is ",batch)
       # Generate papers for the next 20 prompts
       for prompt in batch:
      #   for prompt in prompt1:
           # Print the prompt
           print("prompt is ", prompt)
           response = generate_academic_paper(prompt)
           print("\nGenerated Academic Paper:")
           print("========================\n")
           #print(response)
           for choice in response.choices:
             print(choice.text)
             choice_text_all.append(choice.text)
              # Translate the generated text to Persian
             choice_translated = translate_to_persian(choice.text) #translator.translate(choice.text, dest='fa')
             choice_translated_prompt = translate_to_persian(prompt)#translator.translate(prompt, dest='fa')

             print("\n-----\n Translated is ",choice_translated)
             save_academic_paper("Persian_"+topic,'\n**'+choice_translated_prompt+'**\n'+choice_translated)
             save_academic_paper(topic,'\n**'+prompt+'**\n'+choice.text)
             # Recursively call the function for the nested Completion objects
             if hasattr(choice, 'choices'):
               extract_text(choice)

           #generated_text = response.choices[0].text
           # Print the generated text
           #print(generated_text)
           print("\n end of loop")
           print("========================\n")

           #print("loop")
       # Wait for a short period of time before sending the next batch of prompts
           time.sleep(2)
   return choice_text_all,choice.translated

# Generate academic papers for the given prompts
generate_papers(prompts3)

"""#For solving queta prompt of Openai API;
👇👇🌱
"""

import random
import time
#import openai

from openai import OpenAI

client = OpenAI(
    # defaults to os.environ.get("OPENAI_API_KEY")
    api_key = "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",
)

def retry_with_exponential_backoff(
   func,
   initial_delay: float = 1,
   exponential_base: float = 2,
   jitter: bool = True,
   max_retries: int = 10,
   errors: tuple = (openai.error.RateLimitError,),
):
   """Retry a function with exponential backoff."""
   def wrapper(*args, **kwargs):
       # Initialize variables
       num_retries = 0
       delay = initial_delay
       # Loop until a successful response or max_retries is hit or an exception is raised
       while True:
           try:
               return func(*args, **kwargs)
           # Retry on specified errors
           except errors as e:
               # Increment retries
               num_retries += 1
               # Check if max retries has been reached
               if num_retries > max_retries:
                  raise Exception(
                      f"Maximum number of retries ({max_retries}) exceeded."
                  )
               # Increment the delay
               delay *= exponential_base * (1 + jitter * random.random())
               # Sleep for the delay
               time.sleep(delay)
           # Raise exceptions for any errors not specified
           except Exception as e:
               raise e
   return wrapper

@retry_with_exponential_backoff
def generate_academic_paper_0(prompt_my):
  model_engine = "text-davinci-003"
  max_tokens = 2048

  completion = client.completions.create(
      model="gpt-3.5-turbo-instruct",
      prompt=prompt_my,
      max_tokens=2048,
      temperature=0.3,
      n=1,
      frequency_penalty=0,
  )
  return completion

# Import the necessary libraries and set up the API key
#!pip install openai
import json
#TOPIC =f"strategies for increase the compassion in iranian Women movment and ist relationship with the fatigue of compassion in light triad personality"  #f" the dark triad in psychology and it's relation with {goal}"
topic = f"The Importance of Focusing on Waste Collection in a Waste-Filled World"
import openai
import os
from openai import OpenAI

openai.api_key = "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"# "your_openai_api_key"

client = OpenAI(
    # defaults to os.environ.get("OPENAI_API_KEY")
    api_key="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",
)
#openai.api_key = os.getenv('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')
# Define the prompts
prompts3 = [
f"Find a research topic for a PhD in the area of '{topic}'",
f"Write a detailed proposal on the following research '{topic}'. Make Sure it is free from plagiarism. ",
f"Identify gaps in the literature on '{topic}'",
"Generate 10 academic research questions about Perviuse action",
f"Generate a list of research hypotheses related to '{topic}'"
]



# Function to generate an academic paper

from openai import RateLimitError
from backoff import on_exception, expo

@on_exception(expo, RateLimitError, max_tries=10, max_time=600)
def generate_academic_paper_6(prompt_my):
   model_engine = "text-davinci-003"
   max_tokens = 2048

   completion = client.completions.create(
       model="gpt-3.5-turbo-instruct",
       prompt=prompt_my,
       max_tokens=2048,
       temperature=0.3,
       n=1,
       frequency_penalty=0,
   )
   return completion

choice_text_all=[]

import time
# Function to generate academic papers for given prompts
# Function to generate academic papers for given prompts
def generate_papers(prompts):
   for i in range(0, len(prompts), 20):
       # Slice the prompts list to get the next 20 prompts
       print("I is ",i," Len of prompt Is:", len(prompts))
       batch = prompts[i:i+10]
       print("batch is ",batch)
       # Generate papers for the next 20 prompts
       for prompt in batch:
      #   for prompt in prompt1:
           # Print the prompt
           print("prompt is ", prompt)
           response = generate_academic_paper(prompt)
           print("\nGenerated Academic Paper:")
           print("========================\n")
           #print(response)
           for choice in response.choices:
             print(choice.text)
             choice_text_all.append(choice.text)
             save_academic_paper(TOPIC,'\n**'+prompt+'**\n'+choice.text)
             # Recursively call the function for the nested Completion objects
             if hasattr(choice, 'choices'):
               extract_text(choice)

           #generated_text = response.choices[0].text
           # Print the generated text
           #print(generated_text)
           print("\n end of loop")
           print("========================\n")

           #print("loop")
       # Wait for a short period of time before sending the next batch of prompts
           time.sleep(2)
   return choice.text_all

# Generate academic papers for the given prompts
generate_papers(prompts3)

def render_index_page():
   html_content = """
   <!DOCTYPE html>
   <html>
   <head>
      <title>Generate Academic Papers</title>
   </head>
   <body>
      <h1>Generate Academic Papers</h1>
      <form method="POST">
          <label for="topic">Topic:</label>
          <input type="text" id="topic" name="topic" required>
          <input type="submit" value="Generate">
      </form>
   </body>
   </html>
   """
   return render_template_string(html_content)

def render_result_page(topic, docx_file, pdf_file, choice_text_all):
   html_content = f"""
   <!DOCTYPE html>
   <html>
   <head>
       <title>Result</title>
   </head>
   <body>
       <h1>Result</h1>
       <p>Academic papers have been generated for the topic: {topic}</p>
       <a href="/download/{docx_file}">Download DOCX</a>
       <a href="/download/{pdf_file}">Download PDF</a>
       <h2>Generated Text:</h2>
       <p>{choice_text_all}</p>
   </body>
   </html>
   """
   return render_template_string(html_content)

# app.py

from flask import Flask, request, render_template
from academic_paper_generator import generate_papers

app = Flask(__name__)

@app.route('/', methods=['GET', 'POST'])
def home():
  if request.method == 'POST':
      topic = request.form.get('topic')
      prompts3 = [
          f"Find a research topic for a PhD in the area of '{topic}'",
          f"Write a detailed proposal on the following research '{topic}'. Make Sure it is free from plagiarism. ",
          f"Identify gaps in the literature on '{topic}'",
          "Generate 10 academic research questions about Perviuse action",
          f"Generate a list of research hypotheses related to '{topic}'"
      ]
      generate_papers(prompts3)
      return render_template('result.html', topic=topic)
  else:
      return render_template('index.html')

if __name__ == '__main__':
  app.run(debug=True)