V7

app.py

@@ -3,20 +3,16 @@
 import gradio as gr
 import openai
 import re
-from transformers import pipeline, set_seed
 
 # Set up OpenAI API credentials
 openai.api_key = os.environ["api"]
 
-# Set up Hugging Face pipeline for summarization
-summarizer = pipeline("summarization", model="t5-base", tokenizer="t5-base")
-
 # Define the function that generates the blog article
 def generate_article(topic):
     # Use OpenAI's GPT-3 to generate the article
     prompt = f"Write a blog post about {topic} with 5 different sections."
     response = openai.Completion.create(
-        engine="davinci",
+        engine="text-davinci-002",
         prompt=prompt,
         max_tokens=2048,
         n=1,
@@ -33,16 +29,10 @@ def generate_article(topic):
     section_length = len(article) // 5
     sections = [article[i:i+section_length] for i in range(0, len(article), section_length)]
 
-    # Set seed for reproducibility in Hugging Face spaces
-    set_seed(42)
-
-    # Summarize each section to generate subheadings
-    subheadings = [summarizer(section, max_length=30, min_length=10, do_sample=False)[0]['summary_text'] for section in sections]
-
-    # Combine the sections and subheadings into a formatted blog post
+    # Combine the sections into a formatted blog post
     blog_post = f"# {topic}\n\n"
     for i in range(5):
-        blog_post += f"## {subheadings[i]}\n\n{sections[i]}\n\n"
+        blog_post += f"## Section {i+1}\n\n{sections[i]}\n\n"
     
     return blog_post
 
@@ -52,8 +42,9 @@ iface = gr.Interface(
     inputs=gr.inputs.Textbox("Enter a topic for your blog post"),
     outputs=gr.outputs.HTML(),
     title="Blog Post Generator",
-    description="Generate a blog post on a given topic with 5 different sections.",
+    description="Generate a blog post on a given topic with 5 different sections using OpenAI's GPT-3 text model.",
 )
 
 # Launch the interface
 iface.launch(share=True)
+