Spaces:

benkada
/

aiWeb

Sleeping

App Files Files Community

benkada commited on 5 days ago

Commit

794b7d6

verified ·

1 Parent(s): 1234eb6

Upload 2 files

Browse files

Files changed (2) hide show

index.html +171 -0
main.py +213 -0

index.html ADDED Viewed

	@@ -0,0 +1,171 @@

+<!DOCTYPE html>
+<html lang="en">
+<head>
+  <meta charset="UTF-8" />
+  <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+  <title>AI-Powered Web App</title>
+  <style>
+    :root {
+      --primary-color: #4A90E2;
+      --secondary-color: #f4f6f8;
+      --text-color: #333;
+      --btn-color: #4A90E2;
+      --btn-hover: #357ABD;
+      --card-bg: #fff;
+      --card-shadow: rgba(0, 0, 0, 0.1);
+      --border-radius: 8px;
+      --transition: 0.3s;
+      --max-width: 900px;
+    }
+    * {
+      box-sizing: border-box;
+    }
+    body {
+      font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
+      color: var(--text-color);
+      background: var(--secondary-color);
+      margin: 0;
+      padding: 20px;
+      display: flex;
+      justify-content: center;
+    }
+    .container {
+      width: 100%;
+      max-width: var(--max-width);
+    }
+    h1 {
+      text-align: center;
+      color: var(--primary-color);
+      margin-bottom: 40px;
+    }
+    .section {
+      background: var(--card-bg);
+      padding: 20px;
+      margin-bottom: 30px;
+      border-radius: var(--border-radius);
+      box-shadow: 0 2px 8px var(--card-shadow);
+      transition: transform var(--transition);
+    }
+    .section:hover {
+      transform: translateY(-3px);
+    }
+    h2 {
+      margin-top: 0;
+      color: var(--primary-color);
+      border-bottom: 2px solid var(--secondary-color);
+      padding-bottom: 5px;
+    }
+    label {
+      display: block;
+      margin-top: 15px;
+      font-weight: bold;
+    }
+    input[type="file"], input[type="text"], button, textarea {
+      width: 100%;
+      padding: 10px;
+      margin-top: 8px;
+      border: 1px solid #ccc;
+      border-radius: var(--border-radius);
+      font-size: 1rem;
+    }
+    button {
+      background: var(--btn-color);
+      color: #fff;
+      border: none;
+      margin-top: 20px;
+      padding: 12px;
+      font-size: 1rem;
+      border-radius: var(--border-radius);
+      cursor: pointer;
+      transition: background var(--transition);
+    }
+    button:hover {
+      background: var(--btn-hover);
+    }
+    .results {
+      margin-top: 20px;
+      padding: 15px;
+      background: var(--secondary-color);
+      border-radius: var(--border-radius);
+      border: 1px solid #ddd;
+      min-height: 60px;
+    }
+    @media (min-width: 768px) {
+      .grid {
+        display: flex;
+        gap: 20px;
+      }
+      .grid > div {
+        flex: 1;
+      }
+    }
+  </style>
+</head>
+<body>
+  <div class="container">
+    <h1>AI-Powered Web Application</h1>
+    <!-- Function 1 & 2 in grid on larger screens -->
+    <div class="grid">
+      <!-- Function 1: Document & Image Analysis -->
+      <div class="section" id="analysis-section">
+        <h2>1. Document & Image Analysis</h2>
+        <!-- Text Summarization -->
+        <label for="doc-input-summarize">Upload Document (PDF, DOCX, PPTX, XLSX):</label>
+        <input type="file" id="doc-input-summarize" accept=".pdf,.docx,.pptx,.xlsx" />
+        <button id="summarize-btn">Summarize Document</button>
+        <div class="results" id="summary-result">Your summary will appear here.</div>
+        <!-- Image Captioning -->
+        <label for="img-input-caption">Upload Image (JPG, PNG):</label>
+        <input type="file" id="img-input-caption" accept="image/*" />
+        <button id="caption-btn">Generate Caption</button>
+        <div class="results" id="caption-result">Image caption will appear here.</div>
+      </div>
+      <!-- Function 2: Intelligent Question Answering -->
+      <div class="section" id="qa-section">
+        <h2>2. Intelligent Question Answering</h2>
+        <label for="file-input-qa">Upload Document or Image:</label>
+        <input type="file" id="file-input-qa" accept=".pdf,.docx,.pptx,.xlsx,image/*" />
+        <label for="question-input">Enter Your Question:</label>
+        <input type="text" id="question-input" placeholder="Type your question here..." />
+        <button id="qa-btn">Ask Question</button>
+        <div class="results" id="qa-result">Answer will appear here.</div>
+      </div>
+    </div>
+    <script>
+      async function sendData(url, fileInput, extraData, resultContainer) {
+        const file = fileInput.files[0];
+        if (!file) {
+          resultContainer.textContent = 'Please select a file.';
+          return;
+        }
+        const formData = new FormData();
+        formData.append('file', file);
+        if (extraData) Object.keys(extraData).forEach(key => formData.append(key, extraData[key]));
+        resultContainer.textContent = 'Processing...';
+        try {
+          const res = await fetch(url, { method: 'POST', body: formData });
+          const data = await res.json();
+          resultContainer.textContent = data.result || JSON.stringify(data);
+        } catch (err) {
+          resultContainer.textContent = 'Error: ' + err.message;
+        }
+      }
+      document.getElementById('summarize-btn').addEventListener('click', () => {
+        sendData('/api/summarize', document.getElementById('doc-input-summarize'), null, document.getElementById('summary-result'));
+      });
+      document.getElementById('caption-btn').addEventListener('click', () => {
+        sendData('/api/caption', document.getElementById('img-input-caption'), null, document.getElementById('caption-result'));
+      });
+      document.getElementById('qa-btn').addEventListener('click', () => {
+        const q = document.getElementById('question-input').value.trim();
+        if (!q) { document.getElementById('qa-result').textContent = 'Please enter a question.'; return; }
+        sendData('/api/qa', document.getElementById('file-input-qa'), { question: q }, document.getElementById('qa-result'));
+      });
+    </script>
+  </div>
+</body>
+</html>

main.py ADDED Viewed

	@@ -0,0 +1,213 @@

+from fastapi import FastAPI, UploadFile, File, Form, HTTPException
+from fastapi.responses import JSONResponse
+from fastapi.middleware.cors import CORSMiddleware
+from pydantic import BaseModel
+from typing import Optional
+import os
+import tempfile
+from transformers import pipeline
+import torch
+from PIL import Image
+import pytesseract
+from langchain.chains import LLMChain
+from langchain.prompts import PromptTemplate
+from langchain_community.llms import HuggingFaceHub
+# Initialize FastAPI app
+app = FastAPI(
+    title="AI-Powered Web Application API",
+    description="API for document analysis, image captioning, and question answering",
+    version="1.0.0"
+)
+# CORS configuration
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+# Initialize AI models (lazy loading)
+summarizer = None
+image_captioner = None
+qa_chain = None
+class SummaryRequest(BaseModel):
+    file: UploadFile = File(...)
+class CaptionRequest(BaseModel):
+    file: UploadFile = File(...)
+class QARequest(BaseModel):
+    file: UploadFile = File(...)
+    question: str = Form(...)
+def initialize_models():
+    """Initialize AI models with optimized prompts"""
+    global summarizer, image_captioner, qa_chain
+    # Document summarization model
+    if summarizer is None:
+        summarizer = pipeline(
+            "summarization",
+            model="facebook/bart-large-cnn",
+            device=0 if torch.cuda.is_available() else -1
+        )
+    # Image captioning model
+    if image_captioner is None:
+        image_captioner = pipeline(
+            "image-to-text",
+            model="nlpconnect/vit-gpt2-image-captioning",
+            device=0 if torch.cuda.is_available() else -1
+        )
+    # Question answering chain
+    if qa_chain is None:
+        llm = HuggingFaceHub(
+            repo_id="google/flan-t5-large",
+            model_kwargs={"temperature": 0.1, "max_length": 512}
+        )
+        qa_prompt = PromptTemplate(
+            input_variables=["document", "question"],
+            template="""
+            Using the provided document, answer the following question precisely.
+            If the answer cannot be determined from the document, respond with
+            'The answer cannot be determined from the provided document.'
+            Question: {question}
+            Rules:
+            1. Provide a concise answer (1-3 sentences maximum)
+            2. When possible, reference the specific section of the document that supports your answer
+            3. Maintain numerical precision when answering quantitative questions
+            4. For comparison questions, highlight both items being compared
+            Document: {document}
+            """
+        )
+        qa_chain = LLMChain(llm=llm, prompt=qa_prompt)
+def extract_text_from_file(file: UploadFile) -> str:
+    """Extract text from various file formats"""
+    # Create a temporary file
+    with tempfile.NamedTemporaryFile(delete=False) as temp_file:
+        temp_file.write(file.file.read())
+        temp_path = temp_file.name
+    try:
+        # PDF, DOCX, PPTX, XLSX would need appropriate libraries here
+        # For simplicity, we'll just read text files in this example
+        if file.filename.endswith('.txt'):
+            with open(temp_path, 'r', encoding='utf-8') as f:
+                return f.read()
+        else:
+            # In a real implementation, use libraries like PyPDF2, python-docx, etc.
+            raise HTTPException(
+                status_code=415,
+                detail="File type not supported in this example implementation"
+            )
+    finally:
+        os.unlink(temp_path)
+@app.post("/api/summarize")
+async def summarize_document(file: UploadFile = File(...)):
+    """Summarize a document"""
+    initialize_models()
+    try:
+        # Extract text from the document
+        document_text = extract_text_from_file(file)
+        # Generate summary with optimized prompt
+        summary = summarizer(
+            document_text,
+            max_length=150,
+            min_length=30,
+            do_sample=False,
+            truncation=True
+        )
+        return JSONResponse(
+            content={"status": "success", "result": summary[0]['summary_text']},
+            status_code=200
+        )
+    except Exception as e:
+        raise HTTPException(
+            status_code=500,
+            detail=f"Error processing document: {str(e)}"
+        )
+@app.post("/api/caption")
+async def generate_image_caption(file: UploadFile = File(...)):
+    """Generate caption for an image"""
+    initialize_models()
+    try:
+        # Save the uploaded image temporarily
+        with tempfile.NamedTemporaryFile(suffix=".jpg", delete=False) as temp_file:
+            temp_file.write(file.file.read())
+            temp_path = temp_file.name
+        # Open the image
+        image = Image.open(temp_path)
+        # Generate caption with optimized prompt
+        caption = image_captioner(
+            image,
+            generate_kwargs={
+                "max_length": 50,
+                "num_beams": 4,
+                "early_stopping": True
+            }
+        )
+        return JSONResponse(
+            content={"status": "success", "result": caption[0]['generated_text']},
+            status_code=200
+        )
+    except Exception as e:
+        raise HTTPException(
+            status_code=500,
+            detail=f"Error processing image: {str(e)}"
+        )
+    finally:
+        if 'temp_path' in locals() and os.path.exists(temp_path):
+            os.unlink(temp_path)
+@app.post("/api/qa")
+async def answer_question(
+    file: UploadFile = File(...),
+    question: str = Form(...)
+):
+    """Answer questions based on document content"""
+    initialize_models()
+    try:
+        # Extract text from the document
+        document_text = extract_text_from_file(file)
+        # Get answer using the QA chain
+        answer = qa_chain.run(document=document_text, question=question)
+        return JSONResponse(
+            content={"status": "success", "result": answer},
+            status_code=200
+        )
+    except Exception as e:
+        raise HTTPException(
+            status_code=500,
+            detail=f"Error processing question: {str(e)}"
+        )
+@app.get("/")
+async def health_check():
+    """Health check endpoint"""
+    return {"status": "healthy", "version": "1.0.0"}
+if __name__ == "__main__":
+    import uvicorn
+    uvicorn.run(app, host="0.0.0.0", port=8000)