|
|
|
|
|
|
|
|
from langchain.prompts import PromptTemplate |
|
|
from langchain.chains import LLMChain |
|
|
from langchain.llms import HuggingFaceEndpoint |
|
|
import streamlit as st |
|
|
|
|
|
|
|
|
llm = HuggingFaceEndpoint( |
|
|
endpoint_url= f"https://api-inference.huggingface.co/models/HuggingFaceH4/zephyr-7b-beta", |
|
|
huggingfacehub_api_token='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', |
|
|
task="text-generation", |
|
|
model_kwargs = { |
|
|
"min_length": 8192, |
|
|
"max_length":8192, |
|
|
"temperature":0.1, |
|
|
"max_new_tokens":4000, |
|
|
"num_return_sequences":1 |
|
|
} |
|
|
) |
|
|
|
|
|
|
|
|
topics_template = PromptTemplate( |
|
|
input_variables = ['keyword'], |
|
|
template = """ |
|
|
I want you to Write a comprehensive article about "{keyword}" covering the following aspects: |
|
|
|
|
|
Introduction: Provide an engaging introduction to the topic, highlighting its relevance and significance. |
|
|
History and Background: Explain the historical context and background of {keyword}. |
|
|
Key Concepts and Terminology: Define important terms and concepts related to {keyword}. |
|
|
Current State of {keyword}: Discuss the current trends, developments, and challenges in the field of {keyword}. |
|
|
Use Cases and Applications: Explore practical applications and use cases of {keyword} in various industries. |
|
|
Benefits and Drawbacks: Highlight the advantages and disadvantages of {keyword}. |
|
|
Future Outlook: Predict the future trends and potential advancements in {keyword}. |
|
|
Conclusion: Summarize the key points and reiterate the importance of {keyword}. |
|
|
|
|
|
Ensure that the article is well-structured, informative, and at least 1500 words long. Use SEO best practices for content optimization. |
|
|
""") |
|
|
|
|
|
|
|
|
st.title("Blog Writer") |
|
|
keyword = st.text_input("Input the keyword you wish to write about") |
|
|
|
|
|
topic_writing_chain = LLMChain(llm= llm, prompt=topics_template, verbose = True) |
|
|
|
|
|
|
|
|
with st.spinner('Wait for it...'): |
|
|
if keyword: |
|
|
topics = topic_writing_chain(keyword) |
|
|
st.write(topics) |
|
|
st.success('') |
