Application pushed to huggingface

.gitignore

@@ -0,0 +1 @@
+venv

Dockerfile

@@ -0,0 +1,18 @@
+FROM python:3.10-slim
+
+RUN useradd -m -u 1000 user
+USER user
+ENV PATH="/home/user/.local/bin:$PATH"
+
+WORKDIR /app
+
+# Copy requirements and install Python dependencies
+COPY --chown=user ./requirements.txt requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy application files
+COPY . .
+
+# Run the application
+COPY --chown=user . /app
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]

README.md

@@ -1,10 +1,80 @@
 ---
-title: Text Analyzer Llm
-emoji: 📊
-colorFrom: green
-colorTo: yellow
-sdk: docker
+title: AI Text Assistant
+emoji: 🤖
+colorFrom: purple
+colorTo: blue
+sdk: gradio
+sdk_version: 4.0.0
+app_file: app.py
 pinned: false
+license: mit
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# AI Text Assistant
+
+An interactive web application for text generation, summarization, and next-word prediction using transformer models.
+
+## Features
+
+- **Text Generation**: Generate creative text continuations using Qwen2.5-0.5B-Instruct model
+- **Text Summarization**: Summarize long texts using BART-large-CNN model
+- **Next Word Prediction**: Get top 10 predictions for the next word with probability scores
+
+## Models Used
+
+- **Text Generation**: [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct)
+- **Summarization**: [facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn)
+
+## Project Structure
+
+```
+LocalInference/
+├── app.py              # Main FastAPI application
+├── requirements.txt    # Python dependencies
+├── static/
+│   ├── css/
+│   │   └── style.css  # UI styles
+│   └── js/
+│       └── app.js     # Client-side JavaScript
+└── templates/
+    └── index.html     # Main HTML interface
+```
+
+## Local Setup
+
+1. **Clone the repository:**
+   ```bash
+   git clone <repository-url>
+   cd LocalInference
+   ```
+
+2. **Install dependencies:**
+   ```bash
+   pip install -r requirements.txt
+   ```
+
+3. **Run the application:**
+   ```bash
+   python app.py
+   ```
+   
+   The application will be accessible at `http://localhost:7860`
+
+## Usage
+
+1. Open the application in your web browser
+2. Choose between "Text Generation" or "Text Summarization" mode
+3. Enter your text in the input field
+4. Adjust max tokens and sampling options as needed
+5. Click "Process" to generate results
+6. Use "Get Next Word Predictions" to see likely next words
+
+## API Endpoints
+
+- `GET /` - Web interface
+- `POST /generate` - Generate or summarize text
+- `POST /predict_next` - Get next word predictions
+
+## License
+
+This project is licensed under the MIT License.

app.py

@@ -0,0 +1,80 @@
+from fastapi import FastAPI
+from fastapi.responses import HTMLResponse
+from fastapi.staticfiles import StaticFiles
+from pydantic import BaseModel
+from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
+import torch
+import uvicorn
+import os
+
+app = FastAPI()
+
+# Load models and tokenizer
+model_name = "Qwen/Qwen2.5-0.5B-Instruct"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name)
+generator_pipe = pipeline("text-generation", model=model_name, tokenizer=tokenizer)
+summarizer_pipe = pipeline("summarization", model="facebook/bart-large-cnn")
+
+app.mount("/static", StaticFiles(directory="static"), name="static")
+
+class GenRequest(BaseModel):
+    text: str
+    max_new_tokens: int = 150
+    do_sample: bool = False
+    mode: str = "generate"  # "generate" or "summarize"
+
+@app.get("/", response_class=HTMLResponse)
+async def read_root():
+    with open("templates/index.html", "r") as f:
+        return f.read()
+
+@app.post("/generate")
+def generate(req: GenRequest):
+    if req.mode == "summarize":
+        # Use summarization pipeline
+        out = summarizer_pipe(
+            req.text,
+            max_length=req.max_new_tokens,
+            min_length=30,
+            do_sample=req.do_sample,
+        )
+        return {"generated_text": out[0]["summary_text"]}
+    else:
+        # Use text generation pipeline
+        out = generator_pipe(
+            req.text,
+            max_new_tokens=req.max_new_tokens,
+            do_sample=req.do_sample,
+            truncation=True,
+            return_full_text=False,
+        )
+        return {"generated_text": out[0]["generated_text"]}
+
+@app.post("/predict_next")
+def predict_next(req: GenRequest):
+    """Get top predictions for next word/token"""
+    inputs = tokenizer(req.text, return_tensors="pt")
+    
+    with torch.no_grad():
+        outputs = model(**inputs)
+        next_token_logits = outputs.logits[0, -1, :]
+        
+        # Get top 10 predictions
+        top_k = 10
+        probs = torch.softmax(next_token_logits, dim=-1)
+        top_probs, top_indices = torch.topk(probs, top_k)
+        
+        predictions = []
+        for prob, idx in zip(top_probs.tolist(), top_indices.tolist()):
+            token = tokenizer.decode([idx])
+            predictions.append({
+                "token": token,
+                "probability": round(prob * 100, 2)
+            })
+    
+    return {"predictions": predictions}
+
+if __name__ == "__main__":
+    port = int(os.environ.get("PORT", 7860))
+    uvicorn.run(app, host="0.0.0.0", port=port)

requirements.txt

@@ -0,0 +1,9 @@
+fastapi==0.104.1
+uvicorn[standard]==0.24.0
+pydantic==2.4.2
+transformers==4.45.0
+torch==2.2.0
+numpy<2.0.0
+accelerate==0.24.1
+sentencepiece==0.1.99
+protobuf==4.25.0

static/css/style.css

@@ -0,0 +1,296 @@
+* {
+    margin: 0;
+    padding: 0;
+    box-sizing: border-box;
+}
+
+body {
+    font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, Oxygen, Ubuntu, Cantarell, sans-serif;
+    background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+    min-height: 100vh;
+    padding: 20px;
+}
+
+.container {
+    max-width: 800px;
+    margin: 0 auto;
+    background: white;
+    border-radius: 16px;
+    padding: 40px;
+    box-shadow: 0 20px 60px rgba(0, 0, 0, 0.3);
+}
+
+h1 {
+    color: #333;
+    margin-bottom: 30px;
+    text-align: center;
+    font-size: 2.5em;
+}
+
+h2 {
+    color: #555;
+    margin-bottom: 15px;
+    font-size: 1.3em;
+}
+
+/* Mode Toggle Styles */
+.mode-toggle {
+    display: flex;
+    gap: 10px;
+    margin-bottom: 30px;
+    padding: 5px;
+    background: #f0f0f0;
+    border-radius: 12px;
+}
+
+.toggle-option {
+    flex: 1;
+    text-align: center;
+    cursor: pointer;
+    transition: all 0.3s ease;
+}
+
+.toggle-option input[type="radio"] {
+    display: none;
+}
+
+.toggle-label {
+    display: block;
+    padding: 12px 20px;
+    border-radius: 8px;
+    font-weight: 600;
+    color: #666;
+    transition: all 0.3s ease;
+}
+
+.toggle-option input[type="radio"]:checked + .toggle-label {
+    background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+    color: white;
+    box-shadow: 0 4px 12px rgba(102, 126, 234, 0.4);
+}
+
+.toggle-option:hover .toggle-label {
+    background: #e0e0e0;
+}
+
+.toggle-option input[type="radio"]:checked + .toggle-label:hover {
+    background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+}
+
+/* Input Section */
+.input-section {
+    margin-bottom: 30px;
+}
+
+label {
+    display: block;
+    margin-bottom: 8px;
+    color: #555;
+    font-weight: 600;
+}
+
+textarea {
+    width: 100%;
+    padding: 15px;
+    border: 2px solid #ddd;
+    border-radius: 8px;
+    font-size: 16px;
+    font-family: inherit;
+    resize: vertical;
+    transition: border-color 0.3s ease;
+}
+
+textarea:focus {
+    outline: none;
+    border-color: #667eea;
+}
+
+.controls {
+    display: flex;
+    gap: 20px;
+    margin: 20px 0;
+    flex-wrap: wrap;
+}
+
+.control-group {
+    flex: 1;
+    min-width: 200px;
+}
+
+.control-group input[type="number"] {
+    width: 100%;
+    padding: 10px;
+    border: 2px solid #ddd;
+    border-radius: 8px;
+    font-size: 16px;
+    transition: border-color 0.3s ease;
+}
+
+.control-group input[type="number"]:focus {
+    outline: none;
+    border-color: #667eea;
+}
+
+.control-group input[type="checkbox"] {
+    margin-right: 8px;
+    width: 18px;
+    height: 18px;
+    cursor: pointer;
+}
+
+button {
+    width: 100%;
+    padding: 15px;
+    background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+    color: white;
+    border: none;
+    border-radius: 8px;
+    font-size: 18px;
+    font-weight: 600;
+    cursor: pointer;
+    transition: transform 0.2s ease, box-shadow 0.3s ease;
+}
+
+button:hover {
+    transform: translateY(-2px);
+    box-shadow: 0 8px 20px rgba(102, 126, 234, 0.4);
+}
+
+button:active {
+    transform: translateY(0);
+}
+
+/* Output Section */
+.output-section {
+    margin-top: 30px;
+}
+
+.output-box {
+    background: #f8f9fa;
+    border: 2px solid #ddd;
+    border-radius: 8px;
+    padding: 20px;
+    min-height: 150px;
+    color: #333;
+    line-height: 1.6;
+    white-space: pre-wrap;
+    word-wrap: break-word;
+}
+
+.output-box.loading {
+    color: #999;
+    font-style: italic;
+}
+
+/* Predictions Section */
+.predictions-section {
+    margin-top: 30px;
+    padding-top: 30px;
+    border-top: 2px solid #eee;
+}
+
+.predictions-section button {
+    margin-bottom: 20px;
+}
+
+.predictions-box {
+    background: #f8f9fa;
+    border: 2px solid #ddd;
+    border-radius: 8px;
+    padding: 20px;
+    min-height: 100px;
+    color: #555;
+}
+
+.predictions-list {
+    display: flex;
+    flex-direction: column;
+    gap: 12px;
+}
+
+.prediction-item {
+    background: white;
+    border-radius: 8px;
+    padding: 12px;
+    border: 1px solid #e0e0e0;
+    transition: all 0.2s ease;
+}
+
+.prediction-item:hover {
+    box-shadow: 0 2px 8px rgba(0, 0, 0, 0.1);
+    transform: translateX(4px);
+}
+
+.prediction-header {
+    display: flex;
+    justify-content: space-between;
+    align-items: center;
+    margin-bottom: 8px;
+    gap: 10px;
+}
+
+.prediction-rank {
+    background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+    color: white;
+    padding: 4px 10px;
+    border-radius: 12px;
+    font-size: 0.85em;
+    font-weight: 600;
+    min-width: 35px;
+    text-align: center;
+}
+
+.prediction-token {
+    flex: 1;
+    font-family: 'Courier New', monospace;
+    font-weight: 600;
+    color: #333;
+    font-size: 1.1em;
+}
+
+.prediction-probability {
+    color: #667eea;
+    font-weight: 700;
+    font-size: 1em;
+}
+
+.prediction-bar-container {
+    width: 100%;
+    height: 6px;
+    background: #e0e0e0;
+    border-radius: 3px;
+    overflow: hidden;
+}
+
+.prediction-bar {
+    height: 100%;
+    background: linear-gradient(90deg, #667eea 0%, #764ba2 100%);
+    border-radius: 3px;
+    transition: width 0.5s ease;
+}
+
+@media (max-width: 600px) {
+    .container {
+        padding: 20px;
+    }
+    
+    h1 {
+        font-size: 1.8em;
+    }
+    
+    .controls {
+        flex-direction: column;
+    }
+    
+    .mode-toggle {
+        flex-direction: column;
+    }
+    
+    .prediction-header {
+        flex-wrap: wrap;
+    }
+    
+    .prediction-token {
+        font-size: 0.95em;
+    }
+}

static/js/app.js

@@ -0,0 +1,125 @@
+const generateText = async () => {
+    const textInput = document.getElementById("textInput").value;
+    const maxTokens = document.getElementById("maxTokens").value;
+    const doSample = document.getElementById("doSample").checked;
+    const mode = document.querySelector('input[name="mode"]:checked').value;
+
+    // Show loading state
+    const outputElement = document.getElementById("output");
+    outputElement.innerText = "Processing...";
+    outputElement.classList.add("loading");
+
+    try {
+        const response = await fetch("/generate", {
+            method: "POST",
+            headers: {
+                "Content-Type": "application/json",
+            },
+            body: JSON.stringify({
+                text: textInput,
+                max_new_tokens: parseInt(maxTokens),
+                do_sample: doSample,
+                mode: mode,
+            }),
+        });
+
+        if (response.ok) {
+            const data = await response.json();
+            outputElement.innerText = data.generated_text;
+            outputElement.classList.remove("loading");
+        } else {
+            outputElement.innerText = "Error: Unable to process request.";
+            outputElement.classList.remove("loading");
+        }
+    } catch (error) {
+        outputElement.innerText = "Error: " + error.message;
+        outputElement.classList.remove("loading");
+    }
+};
+
+const predictNext = async () => {
+    const textInput = document.getElementById("textInput").value;
+    
+    if (!textInput.trim()) {
+        alert("Please enter some text first!");
+        return;
+    }
+
+    // Show loading state
+    const predictionsElement = document.getElementById("predictions");
+    predictionsElement.innerHTML = "Loading predictions...";
+
+    try {
+        const response = await fetch("/predict_next", {
+            method: "POST",
+            headers: {
+                "Content-Type": "application/json",
+            },
+            body: JSON.stringify({
+                text: textInput,
+            }),
+        });
+
+        if (response.ok) {
+            const data = await response.json();
+            displayPredictions(data.predictions);
+        } else {
+            predictionsElement.innerHTML = "Error: Unable to get predictions.";
+        }
+    } catch (error) {
+        predictionsElement.innerHTML = "Error: " + error.message;
+    }
+};
+
+const displayPredictions = (predictions) => {
+    const predictionsElement = document.getElementById("predictions");
+    
+    if (predictions.length === 0) {
+        predictionsElement.innerHTML = "No predictions available.";
+        return;
+    }
+
+    let html = '<div class="predictions-list">';
+    predictions.forEach((pred, index) => {
+        const barWidth = pred.probability;
+        html += `
+            <div class="prediction-item">
+                <div class="prediction-header">
+                    <span class="prediction-rank">#${index + 1}</span>
+                    <span class="prediction-token">"${pred.token}"</span>
+                    <span class="prediction-probability">${pred.probability}%</span>
+                </div>
+                <div class="prediction-bar-container">
+                    <div class="prediction-bar" style="width: ${barWidth}%"></div>
+                </div>
+            </div>
+        `;
+    });
+    html += '</div>';
+    
+    predictionsElement.innerHTML = html;
+};
+
+// Update UI based on selected mode
+const updateModeUI = () => {
+    const mode = document.querySelector('input[name="mode"]:checked').value;
+    const placeholder = document.getElementById("textInput");
+    const label = document.querySelector('label[for="textInput"]');
+    
+    if (mode === "summarize") {
+        placeholder.placeholder = "Enter text to summarize...";
+        label.innerText = "Text to Summarize:";
+    } else {
+        placeholder.placeholder = "Enter your prompt...";
+        label.innerText = "Your Prompt:";
+    }
+};
+
+document.getElementById("generateButton").addEventListener("click", generateText);
+document.getElementById("predictButton").addEventListener("click", predictNext);
+document.querySelectorAll('input[name="mode"]').forEach(radio => {
+    radio.addEventListener("change", updateModeUI);
+});
+
+// Initialize UI
+updateModeUI();

templates/index.html

@@ -0,0 +1,62 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>AI Text Tool</title>
+    <link rel="stylesheet" href="/static/css/style.css">
+</head>
+<body>
+    <div class="container">
+        <h1>AI Text Assistant</h1>
+        
+        <!-- Mode Toggle -->
+        <div class="mode-toggle">
+            <label class="toggle-option">
+                <input type="radio" name="mode" value="generate" checked>
+                <span class="toggle-label">Text Generation</span>
+            </label>
+            <label class="toggle-option">
+                <input type="radio" name="mode" value="summarize">
+                <span class="toggle-label">Text Summarization</span>
+            </label>
+        </div>
+
+        <div class="input-section">
+            <label for="textInput">Your Prompt:</label>
+            <textarea id="textInput" rows="6" placeholder="Enter your prompt..."></textarea>
+            
+            <div class="controls">
+                <div class="control-group">
+                    <label for="maxTokens">Max Tokens:</label>
+                    <input type="number" id="maxTokens" value="150" min="10" max="500">
+                </div>
+                
+                <div class="control-group">
+                    <label for="doSample">
+                        <input type="checkbox" id="doSample">
+                        Enable Sampling
+                    </label>
+                </div>
+            </div>
+            
+            <button id="generateButton">Process</button>
+        </div>
+        
+        <div class="output-section">
+            <h2>Output:</h2>
+            <div id="output" class="output-box">Your result will appear here...</div>
+        </div>
+
+        <div class="predictions-section">
+            <h2>Next Word Predictions:</h2>
+            <button id="predictButton">Get Next Word Predictions</button>
+            <div id="predictions" class="predictions-box">
+                Click the button above to see possible next words...
+            </div>
+        </div>
+    </div>
+    
+    <script src="/static/js/app.js"></script>
+</body>
+</html>