Upload 5 files

Dockerfile

@@ -0,0 +1,28 @@
+FROM python:3.10-slim
+
+WORKDIR /app
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    build-essential \
+    && rm -rf /var/lib/apt/lists/*
+
+# Copy requirements first for better caching
+COPY fastapi_app/requirements.txt /app/
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy application files
+COPY fastapi_app /app/
+
+# Create non-root user
+RUN useradd -m -u 1000 appuser && chown -R appuser:appuser /app
+USER appuser
+
+EXPOSE 8000
+
+# Health check using urllib (built-in)
+HEALTHCHECK --interval=30s --timeout=10s --start-period=40s --retries=3 \
+    CMD python -c "import urllib.request; urllib.request.urlopen('http://localhost:8000/health')" || exit 1
+
+# Run the application
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "8000"]

fastapi_app/app.py

@@ -0,0 +1,112 @@
+import torch
+import logging
+from get_model import download_model_from_s3
+from contextlib import asynccontextmanager
+from fastapi import FastAPI, Request, Form
+from fastapi.responses import HTMLResponse
+from fastapi.templating import Jinja2Templates
+from fastapi.staticfiles import StaticFiles
+from transformers import AutoModelForSequenceClassification, AutoTokenizer
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+model = None
+tokenizer = None
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    """Load model on startup and cleanup on shutdown"""
+    global model, tokenizer
+
+    try:
+        logger.info("Starting model download from S3...")
+        model_dir = download_model_from_s3(local_dir="./model")
+
+        logger.info("Loading tokenizer...")
+        tokenizer = AutoTokenizer.from_pretrained(model_dir)
+
+        logger.info("Loading model...")
+        model = AutoModelForSequenceClassification.from_pretrained(model_dir)
+        model.to(device)
+        model.eval()
+
+        logger.info(f"Model loaded successfully on {device}")
+    except Exception as e:
+        logger.error(f"Error loading model: {e}")
+        raise
+
+    yield
+
+    logger.info("Shutting down...")
+
+
+app = FastAPI(title="Sentiment Analysis API", lifespan=lifespan)
+
+templates = Jinja2Templates(directory="templates")
+
+
+@app.get("/", response_class=HTMLResponse)
+async def home(request: Request):
+    """Render the home page"""
+    return templates.TemplateResponse("index.html", {"request": request})
+
+
+@app.post("/predict")
+async def predict(request: Request, text: str = Form(...)):
+    """Predict sentiment for the given text"""
+    if not text.strip():
+        return templates.TemplateResponse(
+            "index.html",
+            {"request": request, "error": "Please enter some text to analyze"},
+        )
+
+    try:
+        inputs = tokenizer(
+            text, return_tensors="pt", truncation=True, max_length=512, padding=True
+        )
+        inputs = {k: v.to(device) for k, v in inputs.items()}
+
+        with torch.no_grad():
+            outputs = model(**inputs)
+            logits = outputs.logits
+            probabilities = torch.nn.functional.softmax(logits, dim=-1)
+            predicted_class = torch.argmax(probabilities, dim=-1).item()
+            confidence = probabilities[0][predicted_class].item()
+
+        sentiment_map = {0: "Negative", 1: "Positive"}
+        sentiment = sentiment_map.get(predicted_class, "Unknown")
+
+        return templates.TemplateResponse(
+            "index.html",
+            {
+                "request": request,
+                "text": text,
+                "sentiment": sentiment,
+                "confidence": round(confidence * 100, 2),
+            },
+        )
+
+    except Exception as e:
+        logger.error(f"Prediction error: {e}")
+        return templates.TemplateResponse(
+            "index.html", {"request": request, "error": f"An error occurred: {str(e)}"}
+        )
+
+
+@app.get("/health")
+async def health_check():
+    """Health check endpoint"""
+    return {
+        "status": "healthy",
+        "model_loaded": model is not None,
+        "device": str(device),
+    }
+
+
+if __name__ == "__main__":
+    import uvicorn
+
+    uvicorn.run(app, host="0.0.0.0", port=8000)

fastapi_app/get_model.py

@@ -0,0 +1,56 @@
+import boto3
+import logging
+
+import os
+from dotenv import load_dotenv
+
+load_dotenv()
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+
+def download_model_from_s3(
+    local_dir="./model", s3_prefix="ml-models/tinybert-sentiment-analysis"
+):
+    """
+    Download the fine-tuned model from S3 bucket
+    """
+    bucket_name = os.getenv("BUCKET_NAME")
+    if not bucket_name:
+        raise ValueError("BUCKET_NAME not found in .env file")
+
+    os.makedirs(local_dir, exist_ok=True)
+
+    s3_client = boto3.client("s3")
+
+    model_files = [
+        "config.json",
+        "model.safetensors",
+        "special_tokens_map.json",
+        "tokenizer_config.json",
+        "tokenizer.json",
+        "vocab.txt",
+    ]
+
+    logger.info(f"Downloading model from S3 bucket: {bucket_name}/{s3_prefix}")
+
+    for file_name in model_files:
+        try:
+            local_file_path = os.path.join(local_dir, file_name)
+
+            if os.path.exists(local_file_path):
+                logger.info(f"File {file_name} already exists, skipping...")
+                continue
+
+            s3_key = f"{s3_prefix}/{file_name}" if s3_prefix else file_name
+
+            logger.info(f"Downloading {s3_key}...")
+            s3_client.download_file(bucket_name, s3_key, local_file_path)
+            logger.info(f"Successfully downloaded {file_name}")
+        except Exception as e:
+            logger.error(f"Error downloading {file_name}: {e}")
+            raise
+
+    logger.info("Model download completed successfully")
+    return local_dir

fastapi_app/requirements.txt

@@ -0,0 +1,8 @@
+fastapi==0.115.6
+jinja2==3.1.5
+boto3==1.34.149
+python-dotenv==1.0.0
+transformers==4.43.3
+torch==2.3.1
+uvicorn==0.34.0
+python-multipart==0.0.18

fastapi_app/templates/index.html

@@ -0,0 +1,106 @@
+<!doctype html>
+<html lang="en">
+  <head>
+    <meta charset="UTF-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Sentiment Analysis - TinyBERT</title>
+    <script src="https://cdn.tailwindcss.com"></script>
+  </head>
+  <body class="bg-gradient-to-br from-slate-50 to-slate-100 min-h-screen">
+    <div class="container mx-auto px-4 py-12 max-w-3xl">
+      <!-- Header -->
+      <div class="text-center mb-12">
+        <h1 class="text-4xl font-bold text-slate-800 mb-3">
+          Sentiment Analysis
+        </h1>
+        <p class="text-slate-600">
+          Powered by Fine-tuned TinyBERT
+        </p>
+      </div>
+
+      <!-- Main Card -->
+      <div class="bg-white rounded-2xl shadow-xl p-8 mb-6">
+        <form method="post" action="/predict" class="space-y-6">
+          <!-- Text Input -->
+          <div>
+            <label for="text" class="block text-sm font-medium text-slate-700 mb-2">
+              Enter text to analyze
+            </label>
+            <textarea
+              id="text"
+              name="text"
+              rows="5"
+              class="w-full px-4 py-3 border border-slate-300 rounded-lg focus:ring-2 focus:ring-blue-500 focus:border-transparent transition resize-none"
+              placeholder="Type or paste your text here..."
+              required
+            >{% if text %}{{ text }}{% endif %}</textarea>
+          </div>
+
+          <!-- Submit Button -->
+          <button
+            type="submit"
+            class="w-full bg-blue-600 hover:bg-blue-700 text-white font-medium py-3 px-6 rounded-lg transition duration-200 shadow-md hover:shadow-lg"
+          >
+            Analyze Sentiment
+          </button>
+        </form>
+
+        <!-- Error Message -->
+        {% if error %}
+        <div class="mt-6 bg-red-50 border-l-4 border-red-500 p-4 rounded">
+          <p class="text-red-700">{{ error }}</p>
+        </div>
+        {% endif %}
+
+        <!-- Results -->
+        {% if sentiment %}
+        <div class="mt-8 border-t pt-6">
+          <h2 class="text-xl font-semibold text-slate-800 mb-4">Results</h2>
+          
+          <div class="grid grid-cols-2 gap-4">
+            <!-- Sentiment -->
+            <div class="bg-slate-50 rounded-lg p-4">
+              <p class="text-sm text-slate-600 mb-1">Sentiment</p>
+              <div class="flex items-center">
+                <span class="text-2xl mr-2">
+                  {% if sentiment == "Positive" %}
+                    😊
+                  {% else %}
+                    😔
+                  {% endif %}
+                </span>
+                <p class="text-2xl font-bold {% if sentiment == 'Positive' %}text-green-600{% else %}text-red-600{% endif %}">
+                  {{ sentiment }}
+                </p>
+              </div>
+            </div>
+
+            <!-- Confidence -->
+            <div class="bg-slate-50 rounded-lg p-4">
+              <p class="text-sm text-slate-600 mb-1">Confidence</p>
+              <p class="text-2xl font-bold text-blue-600">
+                {{ confidence }}%
+              </p>
+            </div>
+          </div>
+
+          <!-- Confidence Bar -->
+          <div class="mt-4">
+            <div class="w-full bg-slate-200 rounded-full h-3 overflow-hidden">
+              <div
+                class="h-full {% if sentiment == 'Positive' %}bg-green-500{% else %}bg-red-500{% endif %} transition-all duration-500"
+                style="width: {{ confidence }}%"
+              ></div>
+            </div>
+          </div>
+        </div>
+        {% endif %}
+      </div>
+
+      <!-- Footer -->
+      <div class="text-center text-slate-500 text-sm">
+        <p>Fine-tuned TinyBERT model for sentiment classification</p>
+      </div>
+    </div>
+  </body>
+</html>