SweetAlpac4/ta-sdn-ml ? reverse-engineered prompt

Build me a small SDN project for testing video streaming quality in a network. I want a setup that uses a Ryu controller with Mininet, collects basic QoS data like delay, jitter, packet loss, and throughput, and then uses machine learning to classify the network path quality in real time. Please include a simple API for prediction, a way to train and evaluate the models, and a script to run experiments and show the results. The goal is to compare Decision Tree and Random Forest and display which one performs better.

Make it easy to run step by step on Ubuntu with Docker, Python, Mininet, and Open vSwitch, and organize the code so it’s clear where the topology, dataset collection, model training, and results live. If you need current library details, look them up online and keep the implementation practical and runnable.