LC150358

import pickle
import gzip
import numpy as np
from network2plus import Network
import matplotlib.pyplot as plt

# ── CARICAMENTO DATASET ──────────────────────────────────────
def load_data():
    with gzip.open('mnist.pkl.gz', 'rb') as f:
        training_data, validation_data, test_data = pickle.load(f, encoding='latin1')

LC150358

#### Libraries
# Standard library
import json
import random
import sys

# Third-party libraries
import numpy as np

LC150358

import pickle
import gzip
import numpy as np
from network2 import Network
import matplotlib.pyplot as plt  # ← AGGIUNGI QUESTO!

def load_data():
    with gzip.open('mnist.pkl.gz', 'rb') as f:
        training_data, validation_data, test_data = pickle.load(f, encoding='latin1')
    # Preprocess training data

LC150358

"""network2.py
~~~~~~~~~~~~~~

An improved version of network.py, implementing the stochastic
gradient descent learning algorithm for a feedforward neural network.
Improvements include the addition of the cross-entropy cost function,
regularization, and better initialization of network weights.  Note
that I have focused on making the code simple, easily readable, and
easily modifiable.  It is not optimized, and omits many desirable
features.

LC150358

import math as mt
import random as rd
 
# seed fisso: garantisce risultati identici
# ad ogni esecuzione (riproducibilità)
rd.seed(1)
 
# ── FUNZIONE SIGMOIDE ──────────────────────────────────────
# trasforma qualsiasi numero in un valore tra 0 e 1
def sigmoide(t):

LC150358

LAB 1 — Classificatore Husky vs Lupo

import math

# ============================================================
# LAB 1 — Classificatore Husky vs Lupo
# rete con 1 neurone, 2 ingressi, funzione sigmoide
# etichette: 0 = Husky,  1 = Lupo
# ============================================================