Finetuned.py

from langchain.prompts import PromptTemplate
from langchain.prompts.few_shot import FewShotPromptTemplate
from langchain.prompts.example_selector import SemanticSimilarityExampleSelector
from langchain.vectorstores import Chroma
from langchain.embeddings import OpenAIEmbeddings

#Few Shot Prompts
# The few shot prompts will guide the algorithm to craft a catchy political headline.
headline_examples = [
  {
    "question": "Can you write me a political headline?",
    "answer":
"""
topic: Politics Unveiled: A Closer Look at the Game of Power and Influence
"""
  },
  {
    "question": "Can you write me a political headline?",
    "answer":
"""
topic: Political Turmoil: Challenges and Controversies Shape the World of Politics
"""
  },
  {
    "question": "Can you write me a political headline?",
    "answer":
"""
topic: Political Landscape Shifts as New Policies and Candidates Emerge
"""
  },
  {
    "question": "Can you write me a political headline?",
    "answer":
"""
topic: Politics: Shaping Society and Nation Through Governance and Policy-Making
"""
  },
  {
    "question": "Can you write me a political headline?",
    "answer":
"""
topic: Breaking News: Political Turmoil Shakes the Nation
"""
  },
   {
    "question": "Can you write me a political headline?",
    "answer":
"""
topic: Political Turmoil Grips Nation as Controversial Policies Divide Citizen
"""
  },
  {
    "question": "Can you write me a political headline?",
    "answer":
"""
topic: Political Landscape Shifts as New Policies Take Center Stage
"""
  },
]

example_selector = SemanticSimilarityExampleSelector.from_examples(
    # This is the list of examples available to select from.
    headline_examples,
    # This is the embedding class used to produce embeddings which are used to measure semantic similarity.
    OpenAIEmbeddings(),
    # This is the VectorStore class that is used to store the embeddings and do a similarity search over.
    Chroma,
    # This is the number of examples to produce.
    k=1
)

#Prompt Templates
#The prompt templates will determine the app's output.
#Prompt Templates for Enhanced Outputs
headline_template = PromptTemplate(
    input_variables = ["question","answer"],
    template = 'question: {question} \n {answer}'
)

headline_prompt = FewShotPromptTemplate(
    example_selector=example_selector,
    example_prompt=headline_template,
    suffix="Prompt: {topic}",
    input_variables=["topic"]

)

press_template = PromptTemplate(
    input_variables = ["headline", "wikipedia_research","google"],
    template = 'I want you to act as a politician. You will research and analyze cultural, economic, political, and social events in the past, collect data from primary sources and use it to develop a press release about what happened during various periods of history. My first suggestion request is: write me a press release based on this headline: {headline} while leveraging this wikipedia research: {wikipedia_research} and google research: {google}'
)

twitter_template = PromptTemplate(
    input_variables = ["press_release"],
    template = 'write me a twitter post based on this press release written by a politician: {press_release}'
)

facebook_template = PromptTemplate(
    input_variables = ["twitter"],
    template = 'write me a facebook post based on this twitter post tweeted by a politician: {twitter}. Do not use emojis excessively.'
)

instagram_template = PromptTemplate(
    input_variables = ["facebook"],
    template = 'write me an instagram post based on this facebook post written by a politician: {facebook}. Do not use emojis excessively.'
)