split_data.py

import pandas as pd
from sklearn.model_selection import train_test_split
import os

BASE_DIR = "/home/azureuser/packet_loss_project"
DATA_DIR = f"{BASE_DIR}/data"

os.makedirs(DATA_DIR, exist_ok=True)

# ✅ READ from project root
data = pd.read_csv(f"{BASE_DIR}/final_dataset.csv")

# Keep only required columns
data = data[["rtt_avg_ms", "jitter_ms", "throughput_mbps"]]

# 70% Train / 30% Test split
train_df, test_df = train_test_split(
    data,
    test_size=0.30,
    random_state=42
)

# ✅ WRITE into data/ folder
train_df.to_csv(f"{DATA_DIR}/train_dataset.csv", index=False)
test_df.to_csv(f"{DATA_DIR}/test_dataset.csv", index=False)

print("✅ Dataset split completed successfully")
print("Training set:", train_df.shape)
print("Testing set :", test_df.shape)