davidechicco

using Pkg

# ----------------------------------------------------------
# Install missing packages
# ----------------------------------------------------------

function ensure_package(pkg::String)
    try
        Base.eval(Main, Meta.parse("using $pkg"))
        println("$pkg already available")

davidechicco

# ==========================================================
# HDBSCAN hyperparameter optimization using DBCV validation
# ==========================================================

using Pkg

# ----------------------------------------------------------
# Install missing packages
# ----------------------------------------------------------

davidechicco

#
# Code snippet written by Davide Chicco <davidechicco@davidechicco.it> on 25th March 2026
# Tested on Linux Xubuntu 24.04.4 LTS with Python 3.11.9
#
import subprocess
import sys
import time  # <-- Added for timing

# --- Upgrade pip ---
bash_command = sys.executable + " -m pip install --upgrade pip"

davidechicco

#
# Code snippet written by Davide Chicco <davidechicco@davidechicco.it> on 25th March 2026
# Tested on Linux Xubuntu 24.04.4 LTS with Python 3.11.9 
#
import subprocess
import sys
import time  # <-- Added for timing

# --- Upgrade pip ---
bash_command = sys.executable + " -m pip install --upgrade pip"

davidechicco

import pandas as pd
import dbcv

data_file_name = 'DB1.csv'
df = pd.read_csv(data_file_name, header = None)
print("data file name read: ", data_file_name)

clusters_file_name = '../Labels/DB1_786.txt'
clusters = pd.read_csv(clusters_file_name, header = None)
clusters.columns = ['cluster']

davidechicco

# Piece of Python code developed by Davide Chicco on 17th November 2024
# developed for Python 3.11 on Linux Xubuntu 22
# released with creative commons license
# contact: davidechicco@davidechicco.it

import numpy as np
from permetrics import ClusteringMetric
from dbcv import dbcv