File size: 5,535 Bytes
acfcb6f fee0742 acfcb6f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 |
from langchain_groq import ChatGroq
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.output_parsers import StrOutputParser
from dotenv import load_dotenv, find_dotenv
import os, sys, getpass
from groq import Groq
import gradio as gr
_ = load_dotenv(find_dotenv())
groq_api_key = os.getenv("GROQ_API_KEY")
class NLP_tasks_crs:
@classmethod
def translator(cls, text: str="Hello world!", language: str="French", style: str="polite"):
# call our prompt engineering class
prompt = Prompt_engineering_crs(text, language, style)
prompt = prompt.translator_prompt()
# call our LLM_crs class
llm = LLM_crs()
llm = llm.chain_llm()
result = llm.invoke(prompt)
return result
@classmethod
def summarization(cls, text: str):
# call our prompt engineering class
prompt = Prompt_engineering_crs(text)
prompt = prompt.summarization_prompt()
# call our LLM_crs class
llm = LLM_crs()
llm = llm.chain_llm()
result = llm.invoke(prompt)
return result
@classmethod
def translator_summarization(cls, text: str, language: str, style: str="polite"):
# call our prompt engineering class
prompt = Prompt_engineering_crs(text, language, style)
prompt = prompt.translate_summarize_prompt()
# call our LLM_crs class
llm = LLM_crs()
llm = llm.chain_llm()
result = llm.invoke(prompt)
return result
@classmethod
def question_answer(cls, question: str):
# call our prompt engineering class
prompt = Prompt_engineering_crs()
prompt = prompt.question_answer_prompt(question)
# call our LLM_crs class
llm = LLM_crs()
llm = llm.chain_llm()
result = llm.invoke(prompt)
return result
class Prompt_engineering_crs:
def __init__(self, text: str="Hello world.", language: str="English", style: str="calm and respectful"):
self.text= text
self.language= language
self.style=style
def translator_prompt(self):
template = """You are the best expert translator of human languagues. Your role is to
firstly detect the correct language of the user text, which is delimited by three
backticks. Secondly, translate that text into the desired language provided by the user, which is
{language}. May sure to use {style} tone as your style. Finally, make sure to sound like a formal native speaker and provide only the final result without
additional information. Thanks.
text: ```{text}``` """
prompt_template = ChatPromptTemplate.from_template(template)
prompt = prompt_template.format_messages(text=self.text, language=self.language, style=self.style)
return prompt
def summarization_prompt(self):
summary = """ You are the best expert summarizer in the world. Your role is to summarize the user text into the detected language.
The provided text is below and between three backticks. Make sure to keep the right context. Don't forget to give
a title to the result of the summarization. Finally, make sure to sound like a formal native speaker and provide
only the final result without additional information or comments. Thanks.
text: ```{text}```"""
prompt_template = ChatPromptTemplate.from_template(summary)
prompt = prompt_template.format_messages(text=self.text)
return prompt
def translate_summarize_prompt(self, language: str="English"):
template =""" You are the best translator and summarizer in the world. Your first role is to translate
the below text into {language} language. Indeed, use the below style during the
translation. Your second role is to summarize into {language} language, the result of the translation
with clear and concise words and expressions. Furthermore, use little imojis during
the translation. Finally, make sure to sound like a formal native speaker and provide only the final result without
additional information or comments. Thanks.
text: {text}
language: {language}
style: {style} """
prompt_template = ChatPromptTemplate.from_template(template)
prompt = prompt_template.format_messages(text=self.text, language=self.language, style=self.style)
return prompt
@classmethod
def question_answer_prompt(cls, question: str):
template = "You are a master of questions and answers. Here, your role is to answer to any questions from " \
"from the user. If you do not know any questions, please state that you don't know them. Don't " \
"invent answers for the questions you do not know. If you have many answers for a question, provide the accurate" \
"ones to the user. Use a polite style to communicate with the user. Indeed, speak like a native speaker. See below the question of the user. " \
"question: {question}"
prompt_template = ChatPromptTemplate.from_template(template)
prompt = prompt_template.format_messages(question = {question})
return prompt
class LLM_crs:
def __init__(self, model="moonshotai/kimi-k2-instruct-0905"):
self.model = model
def chain_llm(self):
llm = ChatGroq(model = self.model)
parser = StrOutputParser()
chain_llm = llm | parser
return chain_llm
|