For Docker deployment

Author: CPPavithra

Dockerfile

@@ -0,0 +1,28 @@
+# Use an official lightweight Python image
+FROM python:3.12-slim
+
+# Set environment variables (better logging)
+ENV PYTHONUNBUFFERED=1
+
+# Set the working directory inside the container
+WORKDIR /app
+
+# Copy only requirements first (better caching)
+COPY requirements.txt .
+
+# Install dependencies
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Create the models directory (fix missing folder issue)
+RUN mkdir -p /app/models
+
+# Copy application files from the root folder
+COPY app.py /app/
+COPY models/k8s_failure_model.pkl /app/models/k8s_failure_model.pkl
+
+# Expose the FastAPI port
+EXPOSE 8000
+
+# Run the FastAPI app using Uvicorn
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "8000"]
+

app.py

@@ -0,0 +1,37 @@
+import pandas as pd
+import joblib
+from fastapi import FastAPI
+from pydantic import BaseModel
+
+# Load the trained model
+model = joblib.load("models/k8s_failure_model.pkl")
+
+app = FastAPI()
+
+# Define request model
+class PredictionRequest(BaseModel):
+    cpu_usage: float
+    memory_usage: float
+    container_network_receive_bytes_total: float
+    container_network_transmit_bytes_total: float
+    container_fs_usage_bytes: float
+    cpu_usage_avg: float
+    memory_usage_avg: float
+    container_network_receive_bytes_total_avg: float
+    container_network_transmit_bytes_total_avg: float
+    container_fs_usage_bytes_avg: float
+    container_restart_count_avg: float
+
+@app.post("/predict")
+async def predict_failure(data: PredictionRequest):
+    # Convert request data to a DataFrame
+    input_data = pd.DataFrame([data.dict()])
+
+    # Temporary fix: Add a dummy 'target_avg' column
+    input_data["target_avg"] = 0  
+
+    # Make prediction
+    prediction = model.predict(input_data)
+    
+    return {"failure_predicted": "YES" if prediction[0] == 1 else "NO"}
+