Update app.py

app.py

@@ -1,83 +1,68 @@
-# -*- coding: utf-8 -*-
-"""Untitled1.ipynb
-
-Automatically generated by Colab.
-
-Original file is located at
-    https://colab.research.google.com/drive/1Oyv-OC4NLyS4SOfffFLwdI0wqvX7K_q2
-"""
-
 import gradio as gr
 from transformers import pipeline
 from PyPDF2 import PdfReader
 from huggingface_hub import InferenceClient
-from google.colab import userdata
 import requests
 from PIL import Image
 import io
 
 pipe = pipeline("text2text-generation", model="asach/simpleT5-resume-summarization")
-
-reader = PdfReader("/KennethGuillont.pdf")
-text = ""
-for page in reader.pages:
-  text += page.extract_text()
-
-summary = pipe(text, max_length=150, min_length=30)[0]['generated_text']
-summary
-
-my_key = userdata.get('HF')
-
+my_key = "YOUR_HUGGING_FACE_API_KEY"
 client = InferenceClient(api_key=my_key)
 
-model_name= 'meta-llama/Llama-3.2-3B-Instruct'
-
-agent_desc = """
-You are an AI agent helps a user generate a prompt to feed into an AI image
-generation model based on a summary of their resume given to you. The image should depict a rabbit
-within the the career feild related to the summary. encase the image prompt between
-two '---\n' marks, to separate it from the rest of the text.
-"""
-
-print(summary)
-
-messages = [
-    {"role": "user", "content": agent_desc},
-    {"role": "user", "content": summary}
-]
-
-stream = client.chat.completions.create(
-    model=model_name,
-    messages=messages,
-    max_tokens=700,
-    stream=True
-)
-
-response_text =""
-
-for chunk in stream:
-  response_text += chunk.choices[0].delta.content
-
-print(response_text)
-
-print(response_text.replace('.','.\n'))
-
-image_prompt = response_text.split('---\n')[1]
-image_prompt
-
-API_URL = "https://api-inference.huggingface.co/models/CompVis/stable-diffusion-v1-4"
-headers = {"Authorization": f"Bearer {my_key}"}
-
-def query(payload):
-	response = requests.post(API_URL, headers=headers, json=payload)
-	return response.content
-
-image_bytes = query({
-	"inputs": image_prompt,
-})
-# You can access the image with PIL.Image for example
-import io
-from PIL import Image
-image = Image.open(io.BytesIO(image_bytes))
-
-image
+def process_pdf(pdf_file):
+    reader = PdfReader(pdf_file.name)
+    text = ""
+    for page in reader.pages:
+        text += page.extract_text()
+
+    summary = pipe(text, max_length=150, min_length=30)[0]['generated_text']
+    agent_desc = """
+    You are an AI agent helps a user generate a prompt to feed into an AI image
+    generation model based on a summary of their resume given to you. The image should depict a rabbit
+    within the the career field related to the summary. Encapsulate the image prompt between
+    two '---' marks.
+    """
+
+    messages = [
+        {"role": "user", "content": agent_desc},
+        {"role": "user", "content": summary}
+    ]
+
+    response_text = ""
+    stream = client.chat.completions.create(
+        model='meta-llama/Llama-3.2-3B-Instruct',
+        messages=messages,
+        max_tokens=700,
+        stream=True
+    )
+
+    for chunk in stream:
+        response_text += chunk.choices[0].delta.content
+
+    image_prompt = response_text.split('---')[1].strip()
+
+    API_URL = "https://api-inference.huggingface.co/models/CompVis/stable-diffusion-v1-4"
+    headers = {"Authorization": f"Bearer {my_key}"}
+
+    def query(payload):
+        response = requests.post(API_URL, headers=headers, json=payload)
+        return response.content
+
+    image_bytes = query({"inputs": image_prompt})
+    image = Image.open(io.BytesIO(image_bytes))
+
+    return summary, image
+
+pdf_input = gr.inputs.File(label="Upload PDF Resume")
+summary_output = gr.outputs.Textbox(label="Resume Summary")
+image_output = gr.outputs.Image(label="Generated Image")
+
+gr.Interface(
+    fn=process_pdf,
+    inputs=pdf_input,
+    outputs=[summary_output, image_output],
+    title="Resume Summarization and Image Generation",
+    description="Upload your PDF resume to get a summary and a related image of a rabbit.",
+    allow_flagging="never"
+).launch()