uczenie-maszynowe/wyk/06_Problem_nadmiernego_dopasowania.ipynb
2022-11-28 11:52:13 +01:00

1833 lines
573 KiB
Plaintext
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

{
"cells": [
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"### Uczenie maszynowe\n",
"# 6. Problem nadmiernego dopasowania"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"## 6.1. Regresja wielomianowa"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"### Wprowadzenie: wybór cech"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"Niech naszym zadaniem będzie przewidzieć cenę działki o kształcie prostokąta.\n",
"\n",
"Jakie cechy wybrać?"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"source": [
"Możemy wybrać dwie cechy:"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
" * $x_1$ szerokość działki, $x_2$ długość działki:\n",
"$$ h_{\\theta}(\\vec{x}) = \\theta_0 + \\theta_1 x_1 + \\theta_2 x_2 $$"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"source": [
"...albo jedną:"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
" * $x_1$ powierzchnia działki:\n",
"$$ h_{\\theta}(\\vec{x}) = \\theta_0 + \\theta_1 x_1 $$"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"source": [
"Można też zauważyć, że cecha „powierzchnia działki” powstaje przez pomnożenie dwóch innych cech: długości działki i jej szerokości."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"source": [
"**Wniosek:** możemy tworzyć nowe cechy na podstawie innych poprzez wykonywanie na nich różnych operacji matematycznych."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"### Regresja wielomianowa"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"source": [
"W regresji wielomianowej będziemy korzystać z cech, które utworzymy jako potęgi cech wyjściowych."
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [],
"source": [
"# Przydatne importy\n",
"\n",
"import ipywidgets as widgets\n",
"import matplotlib.pyplot as plt\n",
"import numpy as np\n",
"import pandas\n",
"\n",
"%matplotlib inline"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [],
"source": [
"# Przydatne funkcje\n",
"\n",
"\n",
"def cost(theta, X, y):\n",
" \"\"\"Wersja macierzowa funkcji kosztu\"\"\"\n",
" m = len(y)\n",
" J = 1.0 / (2.0 * m) * ((X * theta - y).T * (X * theta - y))\n",
" return J.item()\n",
"\n",
"\n",
"def gradient(theta, X, y):\n",
" \"\"\"Wersja macierzowa gradientu funkcji kosztu\"\"\"\n",
" return 1.0 / len(y) * (X.T * (X * theta - y))\n",
"\n",
"\n",
"def gradient_descent(fJ, fdJ, theta, X, y, alpha=0.1, eps=10**-5):\n",
" \"\"\"Algorytm gradientu prostego (wersja macierzowa)\"\"\"\n",
" current_cost = fJ(theta, X, y)\n",
" logs = [[current_cost, theta]]\n",
" while True:\n",
" theta = theta - alpha * fdJ(theta, X, y)\n",
" current_cost, prev_cost = fJ(theta, X, y), current_cost\n",
" if abs(prev_cost - current_cost) > 10**15:\n",
" print(\"Algorithm does not converge!\")\n",
" break\n",
" if abs(prev_cost - current_cost) <= eps:\n",
" break\n",
" logs.append([current_cost, theta])\n",
" return theta, logs\n",
"\n",
"\n",
"def plot_data(X, y, xlabel, ylabel):\n",
" \"\"\"Wykres danych (wersja macierzowa)\"\"\"\n",
" fig = plt.figure(figsize=(16 * 0.6, 9 * 0.6))\n",
" ax = fig.add_subplot(111)\n",
" fig.subplots_adjust(left=0.1, right=0.9, bottom=0.1, top=0.9)\n",
" ax.scatter([X[:, 1]], [y], c=\"r\", s=50, label=\"Dane\")\n",
"\n",
" ax.set_xlabel(xlabel)\n",
" ax.set_ylabel(ylabel)\n",
" ax.margins(0.05, 0.05)\n",
" plt.ylim(y.min() - 1, y.max() + 1)\n",
" plt.xlim(np.min(X[:, 1]) - 1, np.max(X[:, 1]) + 1)\n",
" return fig\n",
"\n",
"\n",
"def plot_fun(fig, fun, X):\n",
" \"\"\"Wykres funkcji `fun`\"\"\"\n",
" ax = fig.axes[0]\n",
" x0 = np.min(X[:, 1]) - 1.0\n",
" x1 = np.max(X[:, 1]) + 1.0\n",
" Arg = np.arange(x0, x1, 0.1)\n",
" Val = fun(Arg)\n",
" return ax.plot(Arg, Val, linewidth=\"2\")\n"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [],
"source": [
"# Wczytanie danych (mieszkania) przy pomocy biblioteki pandas\n",
"\n",
"alldata = pandas.read_csv(\n",
" \"data_flats.tsv\", header=0, sep=\"\\t\", usecols=[\"price\", \"rooms\", \"sqrMetres\"]\n",
")\n",
"data = np.matrix(alldata[[\"sqrMetres\", \"price\"]])\n",
"\n",
"m, n_plus_1 = data.shape\n",
"n = n_plus_1 - 1\n",
"Xn = data[:, 0:n]\n",
"Xn /= np.amax(Xn, axis=0)\n",
"Xn2 = np.power(Xn, 2)\n",
"Xn2 /= np.amax(Xn2, axis=0)\n",
"Xn3 = np.power(Xn, 3)\n",
"Xn3 /= np.amax(Xn3, axis=0)\n",
"\n",
"X = np.matrix(np.concatenate((np.ones((m, 1)), Xn), axis=1)).reshape(m, n + 1)\n",
"X2 = np.matrix(np.concatenate((np.ones((m, 1)), Xn, Xn2), axis=1)).reshape(m, 2 * n + 1)\n",
"X3 = np.matrix(np.concatenate((np.ones((m, 1)), Xn, Xn2, Xn3), axis=1)).reshape(\n",
" m, 3 * n + 1\n",
")\n",
"y = np.matrix(data[:, -1]).reshape(m, 1)\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"Postać ogólna regresji wielomianowej:\n",
"\n",
"$$ h_{\\theta}(x) = \\sum_{i=0}^{n} \\theta_i x^i $$"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [],
"source": [
"# Funkcja regresji wielomianowej\n",
"\n",
"\n",
"def h_poly(Theta, x):\n",
" \"\"\"Funkcja wielomianowa\"\"\"\n",
" return sum(theta * np.power(x, i) for i, theta in enumerate(Theta.tolist()))\n",
"\n",
"\n",
"def polynomial_regression(theta):\n",
" \"\"\"Funkcja regresji wielomianowej\"\"\"\n",
" return lambda x: h_poly(theta, x)\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"source": [
"Najprostszym przypadkiem regresji wielomianowej jest funkcja kwadratowa:"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"Funkcja kwadratowa:\n",
"\n",
"$$ h_{\\theta}(x) = \\theta_0 + \\theta_1 x + \\theta_2 x^2 $$"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"data": {
"text/plain": [
"[<matplotlib.lines.Line2D at 0x7ff4e8126fb0>]"
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 960x540 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig = plot_data(X2, y, xlabel=\"x\", ylabel=\"y\")\n",
"theta_start = np.matrix([0, 0, 0]).reshape(3, 1)\n",
"theta, logs = gradient_descent(cost, gradient, theta_start, X2, y)\n",
"plot_fun(fig, polynomial_regression(theta), X)\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"source": [
"Innym szczególnym przypadkiem regresji wielomianowej jest funkjca sześcienna:"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"Funkcja sześcienna:\n",
"\n",
"$$ h_{\\theta}(x) = \\theta_0 + \\theta_1 x + \\theta_2 x^2 + \\theta_3 x^3 $$"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[[ 397519.38046962]\n",
" [-841341.14146733]\n",
" [2253713.97125102]\n",
" [-244009.07081946]]\n"
]
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 960x540 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig = plot_data(X3, y, xlabel=\"x\", ylabel=\"y\")\n",
"theta_start = np.matrix([0, 0, 0, 0]).reshape(4, 1)\n",
"theta, _ = gradient_descent(cost, gradient, theta_start, X3, y)\n",
"plot_fun(fig, polynomial_regression(theta), X)\n",
"\n",
"print(theta)\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"Regresję wielomianową można potraktować jako szczególny przypadek regresji liniowej wielu zmiennych:\n",
"\n",
"$$ h_{\\theta}(x) = \\theta_0 + \\theta_1 x + \\theta_2 x^2 + \\theta_3 x^3 $$\n",
"$$ x_1 = x, \\quad x_2 = x^2, \\quad x_3 = x^3, \\quad \\vec{x} = \\left[ \\begin{array}{ccc} x_0 \\\\ x_1 \\\\ x_2 \\end{array} \\right] $$"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"source": [
"(W tym przypadku za kolejne cechy przyjmujemy kolejne potęgi zmiennej $x$)."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"Uwaga praktyczna: przyda się normalizacja cech, szczególnie skalowanie!"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"Do tworzenia cech „pochodnych” możemy używać nie tylko potęgowania, ale też innych operacji matematycznych, np.:\n",
"\n",
"$$ h_{\\theta}(x) = \\theta_0 + \\theta_1 x + \\theta_2 \\sqrt{x} $$\n",
"$$ x_1 = x, \\quad x_2 = \\sqrt{x}, \\quad \\vec{x} = \\left[ \\begin{array}{ccc} x_0 \\\\ x_1 \\end{array} \\right] $$"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"source": [
"Jakie zatem cechy wybrać? Najlepiej dopasować je do konkretnego problemu."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"### Wielomianowa regresja logistyczna\n",
"\n",
"Podobne modyfikacje cech możemy również stosować dla regresji logistycznej."
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [],
"source": [
"def powerme(x1, x2, n):\n",
" \"\"\"Funkcja, która generuje n potęg dla zmiennych x1 i x2 oraz ich iloczynów\"\"\"\n",
" X = []\n",
" for m in range(n + 1):\n",
" for i in range(m + 1):\n",
" X.append(np.multiply(np.power(x1, i), np.power(x2, (m - i))))\n",
" return np.hstack(X)\n"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [
{
"data": {
"text/plain": [
"matrix([[ 1. , 0.36596696, -0.11214686],\n",
" [ 0. , 0.4945305 , 0.47110656],\n",
" [ 0. , 0.70290604, -0.92257983],\n",
" [ 0. , 0.46658862, -0.62269739],\n",
" [ 0. , 0.87939462, -0.11408015],\n",
" [ 0. , -0.331185 , 0.84447667],\n",
" [ 0. , -0.54351701, 0.8851383 ],\n",
" [ 0. , 0.91979241, 0.41607012],\n",
" [ 0. , 0.28011742, 0.61431157],\n",
" [ 0. , 0.94754363, -0.78307311]])"
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Wczytanie danych\n",
"import pandas\n",
"import numpy as np\n",
"\n",
"alldata = pandas.read_csv(\"polynomial_logistic.tsv\", sep=\"\\t\")\n",
"data = np.matrix(alldata)\n",
"\n",
"m, n_plus_1 = data.shape\n",
"n = n_plus_1 - 1\n",
"Xn = data[:, 1:]\n",
"\n",
"Xpl = powerme(data[:, 1], data[:, 2], n)\n",
"Ypl = np.matrix(data[:, 0]).reshape(m, 1)\n",
"\n",
"data[:10]\n"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [],
"source": [
"def plot_data_for_classification(X, Y, xlabel, ylabel):\n",
" \"\"\"Wykres danych (wersja macierzowa)\"\"\"\n",
" fig = plt.figure(figsize=(16 * 0.6, 9 * 0.6))\n",
" ax = fig.add_subplot(111)\n",
" fig.subplots_adjust(left=0.1, right=0.9, bottom=0.1, top=0.9)\n",
" X = X.tolist()\n",
" Y = Y.tolist()\n",
" X1n = [x[1] for x, y in zip(X, Y) if y[0] == 0]\n",
" X1p = [x[1] for x, y in zip(X, Y) if y[0] == 1]\n",
" X2n = [x[2] for x, y in zip(X, Y) if y[0] == 0]\n",
" X2p = [x[2] for x, y in zip(X, Y) if y[0] == 1]\n",
" ax.scatter(X1n, X2n, c=\"r\", marker=\"x\", s=50, label=\"Dane\")\n",
" ax.scatter(X1p, X2p, c=\"g\", marker=\"o\", s=50, label=\"Dane\")\n",
"\n",
" ax.set_xlabel(xlabel)\n",
" ax.set_ylabel(ylabel)\n",
" ax.margins(0.05, 0.05)\n",
" return fig\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"source": [
"Przyjmijmy, że mamy następujące dane i chcemy przeprowadzić klasyfikację dwuklasową dla następujących klas:\n",
" * czerwone krzyżyki\n",
" * zielone kółka"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 960x540 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig = plot_data_for_classification(Xpl, Ypl, xlabel=r\"$x_1$\", ylabel=r\"$x_2$\")\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"Propozycja hipotezy:\n",
"\n",
"$$ h_\\theta(x) = g(\\theta^T x) = g(\\theta_0 + \\theta_1 x_1 + \\theta_2 x_2 + \\theta_3 x_3 + \\theta_4 x_4 + \\theta_5 x_5) \\; , $$\n",
"\n",
"gdzie $g$ funkcja logistyczna, $x_3 = x_1^2$, $x_4 = x_2^2$, $x_5 = x_1 x_2$."
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [],
"source": [
"def safeSigmoid(x, eps=0):\n",
" \"\"\"Funkcja sigmoidalna zmodyfikowana w taki sposób,\n",
" żeby wartości zawsz były odległe od asymptot o co najmniej eps\n",
" \"\"\"\n",
" y = 1.0 / (1.0 + np.exp(-x))\n",
" if eps > 0:\n",
" y[y < eps] = eps\n",
" y[y > 1 - eps] = 1 - eps\n",
" return y\n",
"\n",
"\n",
"def h(theta, X, eps=0.0):\n",
" \"\"\"Funkcja hipotezy\"\"\"\n",
" return safeSigmoid(X * theta, eps)\n",
"\n",
"\n",
"def J(h, theta, X, y, lamb=0):\n",
" \"\"\"Funkcja kosztu\"\"\"\n",
" m = len(y)\n",
" f = h(theta, X, eps=10**-7)\n",
" j = (\n",
" -np.sum(np.multiply(y, np.log(f)) + np.multiply(1 - y, np.log(1 - f)), axis=0)\n",
" / m\n",
" )\n",
" if lamb > 0:\n",
" j += lamb / (2 * m) * np.sum(np.power(theta[1:], 2))\n",
" return j\n",
"\n",
"\n",
"def dJ(h, theta, X, y, lamb=0):\n",
" \"\"\"Pochodna funkcji kosztu\"\"\"\n",
" g = 1.0 / y.shape[0] * (X.T * (h(theta, X) - y))\n",
" if lamb > 0:\n",
" g[1:] += lamb / float(y.shape[0]) * theta[1:]\n",
" return g\n",
"\n",
"\n",
"def classifyBi(theta, X):\n",
" \"\"\"Funkcja decyzji\"\"\"\n",
" prob = h(theta, X)\n",
" return prob\n"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [],
"source": [
"def GD(h, fJ, fdJ, theta, X, y, alpha=0.01, eps=10**-3, maxSteps=10000):\n",
" \"\"\"Metoda gradientu prostego dla regresji logistycznej\"\"\"\n",
" errorCurr = fJ(h, theta, X, y)\n",
" errors = [[errorCurr, theta]]\n",
" while True:\n",
" # oblicz nowe theta\n",
" theta = theta - alpha * fdJ(h, theta, X, y)\n",
" # raportuj poziom błędu\n",
" errorCurr, errorPrev = fJ(h, theta, X, y), errorCurr\n",
" # kryteria stopu\n",
" if abs(errorPrev - errorCurr) <= eps:\n",
" break\n",
" if len(errors) > maxSteps:\n",
" break\n",
" errors.append([errorCurr, theta])\n",
" return theta, errors\n"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"theta = [[ 1.59558981]\n",
" [ 0.12602307]\n",
" [ 0.65718518]\n",
" [-5.26367581]\n",
" [ 1.96832544]\n",
" [-6.97946065]]\n"
]
}
],
"source": [
"# Uruchomienie metody gradientu prostego dla regresji logistycznej\n",
"theta_start = np.matrix(np.zeros(Xpl.shape[1])).reshape(Xpl.shape[1], 1)\n",
"theta, errors = GD(\n",
" h, J, dJ, theta_start, Xpl, Ypl, alpha=0.1, eps=10**-7, maxSteps=10000\n",
")\n",
"print(r\"theta = {}\".format(theta))\n"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [],
"source": [
"def plot_decision_boundary(fig, theta, X):\n",
" \"\"\"Wykres granicy klas\"\"\"\n",
" ax = fig.axes[0]\n",
" xx, yy = np.meshgrid(np.arange(-1.0, 1.0, 0.02), np.arange(-1.0, 1.0, 0.02))\n",
" l = len(xx.ravel())\n",
" C = powerme(xx.reshape(l, 1), yy.reshape(l, 1), n)\n",
" z = classifyBi(theta, C).reshape(int(np.sqrt(l)), int(np.sqrt(l)))\n",
"\n",
" plt.contour(xx, yy, z, levels=[0.5], lw=3)\n"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/tmp/ipykernel_74/1169766636.py:9: UserWarning: The following kwargs were not used by contour: 'lw'\n",
" plt.contour(xx, yy, z, levels=[0.5], lw=3)\n"
]
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 960x540 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig = plot_data_for_classification(Xpl, Ypl, xlabel=r\"$x_1$\", ylabel=r\"$x_2$\")\n",
"plot_decision_boundary(fig, theta, Xpl)\n"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [],
"source": [
"# Wczytanie danych\n",
"\n",
"alldata = pandas.read_csv(\"polynomial_logistic.tsv\", sep=\"\\t\")\n",
"data = np.matrix(alldata)\n",
"\n",
"m, n_plus_1 = data.shape\n",
"Xn = data[:, 1:]\n",
"\n",
"n = 10\n",
"Xpl = powerme(data[:, 1], data[:, 2], n)\n",
"Ypl = np.matrix(data[:, 0]).reshape(m, 1)\n",
"\n",
"theta_start = np.matrix(np.zeros(Xpl.shape[1])).reshape(Xpl.shape[1], 1)\n",
"theta, errors = GD(\n",
" h, J, dJ, theta_start, Xpl, Ypl, alpha=0.1, eps=10**-7, maxSteps=10000\n",
")\n"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/tmp/ipykernel_74/1169766636.py:9: UserWarning: The following kwargs were not used by contour: 'lw'\n",
" plt.contour(xx, yy, z, levels=[0.5], lw=3)\n"
]
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 960x540 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Przykład dla większej liczby cech\n",
"fig = plot_data_for_classification(Xpl, Ypl, xlabel=r\"$x_1$\", ylabel=r\"$x_2$\")\n",
"plot_decision_boundary(fig, theta, Xpl)\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"## 6.2. Problem nadmiernego dopasowania"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"### Obciążenie a wariancja"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [],
"source": [
"# Dane do prostego przykładu\n",
"\n",
"data = np.matrix(\n",
" [\n",
" [0.0, 0.0],\n",
" [0.5, 1.8],\n",
" [1.0, 4.8],\n",
" [1.6, 7.2],\n",
" [2.6, 8.8],\n",
" [3.0, 9.0],\n",
" ]\n",
")\n",
"\n",
"m, n_plus_1 = data.shape\n",
"n = n_plus_1 - 1\n",
"Xn1 = data[:, 0:n]\n",
"Xn1 /= np.amax(Xn1, axis=0)\n",
"Xn2 = np.power(Xn1, 2)\n",
"Xn2 /= np.amax(Xn2, axis=0)\n",
"Xn3 = np.power(Xn1, 3)\n",
"Xn3 /= np.amax(Xn3, axis=0)\n",
"Xn4 = np.power(Xn1, 4)\n",
"Xn4 /= np.amax(Xn4, axis=0)\n",
"Xn5 = np.power(Xn1, 5)\n",
"Xn5 /= np.amax(Xn5, axis=0)\n",
"\n",
"X1 = np.matrix(np.concatenate((np.ones((m, 1)), Xn1), axis=1)).reshape(m, n + 1)\n",
"X2 = np.matrix(np.concatenate((np.ones((m, 1)), Xn1, Xn2), axis=1)).reshape(\n",
" m, 2 * n + 1\n",
")\n",
"X5 = np.matrix(\n",
" np.concatenate((np.ones((m, 1)), Xn1, Xn2, Xn3, Xn4, Xn5), axis=1)\n",
").reshape(m, 5 * n + 1)\n",
"y = np.matrix(data[:, -1]).reshape(m, 1)\n"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 960x540 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig = plot_data(X1, y, xlabel=\"x\", ylabel=\"y\")\n"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"scrolled": true,
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"data": {
"text/plain": [
"[<matplotlib.lines.Line2D at 0x7ff4e7adf4f0>]"
]
},
"execution_count": 20,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 960x540 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig = plot_data(X1, y, xlabel=\"x\", ylabel=\"y\")\n",
"theta_start = np.matrix([0, 0]).reshape(2, 1)\n",
"theta, _ = gradient_descent(cost, gradient, theta_start, X1, y, eps=0.00001)\n",
"plot_fun(fig, polynomial_regression(theta), X1)\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"Ten model ma duże **obciążenie** (**błąd systematyczny**, *bias*) zachodzi **niedostateczne dopasowanie** (*underfitting*)."
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"data": {
"text/plain": [
"[<matplotlib.lines.Line2D at 0x7ff478d03df0>]"
]
},
"execution_count": 21,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 960x540 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig = plot_data(X2, y, xlabel=\"x\", ylabel=\"y\")\n",
"theta_start = np.matrix([0, 0, 0]).reshape(3, 1)\n",
"theta, _ = gradient_descent(cost, gradient, theta_start, X2, y, eps=0.000001)\n",
"plot_fun(fig, polynomial_regression(theta), X1)\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"Ten model jest odpowiednio dopasowany."
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"data": {
"text/plain": [
"[<matplotlib.lines.Line2D at 0x7ff47947b9a0>]"
]
},
"execution_count": 22,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 960x540 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig = plot_data(X5, y, xlabel=\"x\", ylabel=\"y\")\n",
"theta_start = np.matrix([0, 0, 0, 0, 0, 0]).reshape(6, 1)\n",
"theta, _ = gradient_descent(cost, gradient, theta_start, X5, y, alpha=0.5, eps=10**-7)\n",
"plot_fun(fig, polynomial_regression(theta), X1)\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"Ten model ma dużą **wariancję** (*variance*) zachodzi **nadmierne dopasowanie** (*overfitting*)."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"source": [
"(Zwróć uwagę na dziwny kształt krzywej w lewej części wykresu to m.in. efekt nadmiernego dopasowania)."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"source": [
"Nadmierne dopasowanie występuje, gdy model ma zbyt dużo stopni swobody w stosunku do ilości danych wejściowych.\n",
"\n",
"Jest to zjawisko niepożądane.\n",
"\n",
"Możemy obrazowo powiedzieć, że nadmierne dopasowanie występuje, gdy model zaczyna modelować szum/zakłócenia w danych zamiast ich „głównego nurtu”. "
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"source": [
"Zobacz też: https://pl.wikipedia.org/wiki/Nadmierne_dopasowanie"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"<img style=\"margin:auto\" width=\"90%\" src=\"fit.png\"/>"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"### Obciążenie (błąd systematyczny, *bias*)\n",
"\n",
"* Wynika z błędnych założeń co do algorytmu uczącego się.\n",
"* Duże obciążenie powoduje niedostateczne dopasowanie."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"### Wariancja (*variance*)\n",
"\n",
"* Wynika z nadwrażliwości na niewielkie fluktuacje w zbiorze uczącym.\n",
"* Wysoka wariancja może spowodować nadmierne dopasowanie (modelując szum zamiast sygnału)."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"<img style=\"margin:auto\" width=\"40%\" src=\"bias2.png\"/>"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"<img style=\"margin:auto\" width=\"60%\" src=\"curves.jpg\"/>"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"## 6.3. Regularyzacja"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [],
"source": [
"def SGD(\n",
" h,\n",
" fJ,\n",
" fdJ,\n",
" theta,\n",
" X,\n",
" Y,\n",
" alpha=0.001,\n",
" maxEpochs=1.0,\n",
" batchSize=100,\n",
" adaGrad=False,\n",
" logError=False,\n",
" validate=0.0,\n",
" valStep=100,\n",
" lamb=0,\n",
" trainsetsize=1.0,\n",
"):\n",
" \"\"\"Stochastic Gradient Descent - stochastyczna wersja metody gradientu prostego\n",
" (więcej na ten temat na następnym wykładzie)\n",
" \"\"\"\n",
" errorsX, errorsY = [], []\n",
" errorsVX, errorsVY = [], []\n",
"\n",
" XT, YT = X, Y\n",
"\n",
" m_end = int(trainsetsize * len(X))\n",
"\n",
" if validate > 0:\n",
" mv = int(X.shape[0] * validate)\n",
" XV, YV = X[:mv], Y[:mv]\n",
" XT, YT = X[mv:m_end], Y[mv:m_end]\n",
" m, n = XT.shape\n",
"\n",
" start, end = 0, batchSize\n",
" maxSteps = (m * float(maxEpochs)) / batchSize\n",
"\n",
" if adaGrad:\n",
" hgrad = np.matrix(np.zeros(n)).reshape(n, 1)\n",
"\n",
" for i in range(int(maxSteps)):\n",
" XBatch, YBatch = XT[start:end, :], YT[start:end, :]\n",
"\n",
" grad = fdJ(h, theta, XBatch, YBatch, lamb=lamb)\n",
" if adaGrad:\n",
" hgrad += np.multiply(grad, grad)\n",
" Gt = 1.0 / (10**-7 + np.sqrt(hgrad))\n",
" theta = theta - np.multiply(alpha * Gt, grad)\n",
" else:\n",
" theta = theta - alpha * grad\n",
"\n",
" if logError:\n",
" errorsX.append(float(i * batchSize) / m)\n",
" errorsY.append(fJ(h, theta, XBatch, YBatch).item())\n",
" if validate > 0 and i % valStep == 0:\n",
" errorsVX.append(float(i * batchSize) / m)\n",
" errorsVY.append(fJ(h, theta, XV, YV).item())\n",
"\n",
" if start + batchSize < m:\n",
" start += batchSize\n",
" else:\n",
" start = 0\n",
" end = min(start + batchSize, m)\n",
" return theta, (errorsX, errorsY, errorsVX, errorsVY)\n"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [],
"source": [
"# Przygotowanie danych do przykładu regularyzacji\n",
"\n",
"n = 6\n",
"\n",
"data = np.matrix(np.loadtxt(\"ex2data2.txt\", delimiter=\",\"))\n",
"np.random.shuffle(data)\n",
"\n",
"X = powerme(data[:, 0], data[:, 1], n)\n",
"Y = data[:, 2]\n"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [],
"source": [
"def draw_regularization_example(\n",
" X, Y, lamb=0, alpha=1, adaGrad=True, maxEpochs=2500, validate=0.25\n",
"):\n",
" \"\"\"Rusuje przykład regularyzacji\"\"\"\n",
" plt.figure(figsize=(16, 8))\n",
" plt.subplot(121)\n",
" plt.scatter(\n",
" X[:, 2].tolist(),\n",
" X[:, 1].tolist(),\n",
" c=Y.tolist(),\n",
" s=100,\n",
" cmap=plt.cm.get_cmap(\"prism\"),\n",
" )\n",
"\n",
" theta = np.matrix(np.zeros(X.shape[1])).reshape(X.shape[1], 1)\n",
" thetaBest, err = SGD(\n",
" h,\n",
" J,\n",
" dJ,\n",
" theta,\n",
" X,\n",
" Y,\n",
" alpha=alpha,\n",
" adaGrad=adaGrad,\n",
" maxEpochs=maxEpochs,\n",
" batchSize=100,\n",
" logError=True,\n",
" validate=validate,\n",
" valStep=1,\n",
" lamb=lamb,\n",
" )\n",
"\n",
" xx, yy = np.meshgrid(np.arange(-1.5, 1.5, 0.02), np.arange(-1.5, 1.5, 0.02))\n",
" l = len(xx.ravel())\n",
" C = powerme(xx.reshape(l, 1), yy.reshape(l, 1), n)\n",
" z = classifyBi(thetaBest, C).reshape(int(np.sqrt(l)), int(np.sqrt(l)))\n",
"\n",
" plt.contour(xx, yy, z, levels=[0.5], lw=3)\n",
" plt.ylim(-1, 1.2)\n",
" plt.xlim(-1, 1.2)\n",
" plt.legend()\n",
" plt.subplot(122)\n",
" plt.plot(err[0], err[1], lw=3, label=\"Training error\")\n",
" if validate > 0:\n",
" plt.plot(err[2], err[3], lw=3, label=\"Validation error\")\n",
" plt.legend()\n",
" plt.ylim(0.2, 0.8)\n"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {
"scrolled": true,
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/tmp/ipykernel_74/2678993393.py:5: RuntimeWarning: overflow encountered in exp\n",
" y = 1.0 / (1.0 + np.exp(-x))\n",
"/tmp/ipykernel_74/2651435526.py:38: UserWarning: The following kwargs were not used by contour: 'lw'\n",
" plt.contour(xx, yy, z, levels=[0.5], lw=3)\n",
"No artists with labels found to put in legend. Note that artists whose label start with an underscore are ignored when legend() is called with no argument.\n"
]
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1600x800 with 2 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"draw_regularization_example(X, Y)\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"### Regularyzacja\n",
"\n",
"Regularyzacja jest metodą zapobiegania zjawisku nadmiernego dopasowania (*overfitting*) poprzez odpowiednie zmodyfikowanie funkcji kosztu.\n",
"\n",
"Do funkcji kosztu dodawane jest specjalne wyrażenie (**wyrażenie regularyzacyjne** zaznaczone na czerwono w poniższych wzorach), będące „karą” za ekstremalne wartości parametrów $\\theta$.\n",
"\n",
"W ten sposób preferowane są wektory $\\theta$ z mniejszymi wartosciami parametrów mają automatycznie niższy koszt.\n",
"\n",
"Jak silną regularyzację chcemy zastosować? Możemy o tym zadecydować, dobierajac odpowiednio **parametr regularyzacji** $\\lambda$."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"source": [
"Przedstawiona tu metoda regularyzacji to tzw. metoda L2 (*ridge*). Istnieją również inne metody regularyzacji, które charakteryzują się trochę innymi własnościami, np. L2 (*lasso*) lub *elastic net*. Więcej na ten temat można przeczytać np. tu:\n",
"* [L1 and L2 Regularization Methods](https://towardsdatascience.com/l1-and-l2-regularization-methods-ce25e7fc831c)\n",
"* [Ridge and Lasso Regression: L1 and L2 Regularization](https://towardsdatascience.com/ridge-and-lasso-regression-a-complete-guide-with-python-scikit-learn-e20e34bcbf0b)\n",
"* [Elastic Net Regression](https://towardsdatascience.com/elastic-net-regression-from-sklearn-to-tensorflow-3b48eee45e91)"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"### Regularyzacja dla regresji liniowej funkcja kosztu\n",
"\n",
"$$\n",
"J(\\theta) \\, = \\, \\dfrac{1}{2m} \\left( \\displaystyle\\sum_{i=1}^{m} h_\\theta(x^{(i)}) - y^{(i)} \\color{red}{ + \\lambda \\displaystyle\\sum_{j=1}^{n} \\theta^2_j } \\right)\n",
"$$"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"* $\\lambda$ parametr regularyzacji\n",
"* jeżeli $\\lambda$ jest zbyt mały, skutkuje to nadmiernym dopasowaniem\n",
"* jeżeli $\\lambda$ jest zbyt duży, skutkuje to niedostatecznym dopasowaniem"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"### Regularyzacja dla regresji liniowej gradient\n",
"\n",
"$$\\small\n",
"\\begin{array}{llll}\n",
"\\dfrac{\\partial J(\\theta)}{\\partial \\theta_0} &=& \\dfrac{1}{m}\\displaystyle\\sum_{i=1}^m \\left( h_{\\theta}(x^{(i)})-y^{(i)} \\right) x^{(i)}_0 & \\textrm{dla $j = 0$ }\\\\\n",
"\\dfrac{\\partial J(\\theta)}{\\partial \\theta_j} &=& \\dfrac{1}{m}\\displaystyle\\sum_{i=1}^m \\left( h_{\\theta}(x^{(i)})-y^{(i)} \\right) x^{(i)}_j \\color{red}{+ \\dfrac{\\lambda}{m}\\theta_j} & \\textrm{dla $j = 1, 2, \\ldots, n $} \\\\\n",
"\\end{array} \n",
"$$"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"### Regularyzacja dla regresji logistycznej funkcja kosztu\n",
"\n",
"$$\n",
"\\begin{array}{rtl}\n",
"J(\\theta) & = & -\\dfrac{1}{m} \\left( \\displaystyle\\sum_{i=1}^{m} y^{(i)} \\log h_\\theta(x^{(i)}) + \\left( 1-y^{(i)} \\right) \\log \\left( 1-h_\\theta(x^{(i)}) \\right) \\right) \\\\\n",
"& & \\color{red}{ + \\dfrac{\\lambda}{2m} \\displaystyle\\sum_{j=1}^{n} \\theta^2_j } \\\\\n",
"\\end{array}\n",
"$$"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"### Regularyzacja dla regresji logistycznej gradient\n",
"\n",
"$$\\small\n",
"\\begin{array}{llll}\n",
"\\dfrac{\\partial J(\\theta)}{\\partial \\theta_0} &=& \\dfrac{1}{m}\\displaystyle\\sum_{i=1}^m \\left( h_{\\theta}(x^{(i)})-y^{(i)} \\right) x^{(i)}_0 & \\textrm{dla $j = 0$ }\\\\\n",
"\\dfrac{\\partial J(\\theta)}{\\partial \\theta_j} &=& \\dfrac{1}{m}\\displaystyle\\sum_{i=1}^m \\left( h_{\\theta}(x^{(i)})-y^{(i)} \\right) x^{(i)}_j \\color{red}{+ \\dfrac{\\lambda}{m}\\theta_j} & \\textrm{dla $j = 1, 2, \\ldots, n $} \\\\\n",
"\\end{array} \n",
"$$"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"### Implementacja metody regularyzacji"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [],
"source": [
"def J_(h, theta, X, y, lamb=0):\n",
" \"\"\"Funkcja kosztu z regularyzacją\"\"\"\n",
" m = float(len(y))\n",
" f = h(theta, X, eps=10**-7)\n",
" j = 1.0 / m * -np.sum(\n",
" np.multiply(y, np.log(f)) + np.multiply(1 - y, np.log(1 - f)), axis=0\n",
" ) + lamb / (2 * m) * np.sum(np.power(theta[1:], 2))\n",
" return j\n",
"\n",
"\n",
"def dJ_(h, theta, X, y, lamb=0):\n",
" \"\"\"Gradient funkcji kosztu z regularyzacją\"\"\"\n",
" m = float(y.shape[0])\n",
" g = 1.0 / y.shape[0] * (X.T * (h(theta, X) - y))\n",
" g[1:] += lamb / m * theta[1:]\n",
" return g\n"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [],
"source": [
"slider_lambda = widgets.FloatSlider(\n",
" min=0.0, max=0.5, step=0.005, value=0.01, description=r\"$\\lambda$\", width=300\n",
")\n",
"\n",
"\n",
"def slide_regularization_example_2(lamb):\n",
" draw_regularization_example(X, Y, lamb=lamb)\n"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {
"scrolled": false,
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "546e94aba9124fcf8da68fc67ab9cdd0",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"interactive(children=(FloatSlider(value=0.01, description='$\\\\lambda$', max=0.5, step=0.005), Button(descripti…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"<function __main__.slide_regularization_example_2(lamb)>"
]
},
"execution_count": 29,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"widgets.interact_manual(slide_regularization_example_2, lamb=slider_lambda)\n"
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [],
"source": [
"def cost_lambda_fun(lamb):\n",
" \"\"\"Koszt w zależności od parametru regularyzacji lambda\"\"\"\n",
" theta = np.matrix(np.zeros(X.shape[1])).reshape(X.shape[1], 1)\n",
" thetaBest, err = SGD(\n",
" h,\n",
" J,\n",
" dJ,\n",
" theta,\n",
" X,\n",
" Y,\n",
" alpha=1,\n",
" adaGrad=True,\n",
" maxEpochs=2500,\n",
" batchSize=100,\n",
" logError=True,\n",
" validate=0.25,\n",
" valStep=1,\n",
" lamb=lamb,\n",
" )\n",
" return err[1][-1], err[3][-1]\n",
"\n",
"\n",
"def plot_cost_lambda():\n",
" \"\"\"Wykres kosztu w zależności od parametru regularyzacji lambda\"\"\"\n",
" plt.figure(figsize=(16, 8))\n",
" ax = plt.subplot(111)\n",
" Lambda = np.arange(0.0, 1.0, 0.01)\n",
" Costs = [cost_lambda_fun(lamb) for lamb in Lambda]\n",
" CostTrain = [cost[0] for cost in Costs]\n",
" CostCV = [cost[1] for cost in Costs]\n",
" plt.plot(Lambda, CostTrain, lw=3, label=\"training error\")\n",
" plt.plot(Lambda, CostCV, lw=3, label=\"validation error\")\n",
" ax.set_xlabel(r\"$\\lambda$\")\n",
" ax.set_ylabel(\"cost\")\n",
" plt.legend()\n",
" plt.ylim(0.2, 0.8)\n"
]
},
{
"cell_type": "code",
"execution_count": 31,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1600x800 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plot_cost_lambda()\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"## 5.4. Krzywa uczenia się"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"* Krzywa uczenia pozwala sprawdzić, czy uczenie przebiega poprawnie.\n",
"* Krzywa uczenia to wykres zależności między wielkością zbioru treningowego a wartością funkcji kosztu.\n",
"* Wraz ze wzrostem wielkości zbioru treningowego wartość funkcji kosztu na zbiorze treningowym rośnie.\n",
"* Wraz ze wzrostem wielkości zbioru treningowego wartość funkcji kosztu na zbiorze walidacyjnym maleje."
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {
"slideshow": {
"slide_type": "notes"
}
},
"outputs": [],
"source": [
"def cost_trainsetsize_fun(m):\n",
" \"\"\"Koszt w zależności od wielkości zbioru uczącego\"\"\"\n",
" theta = np.matrix(np.zeros(X.shape[1])).reshape(X.shape[1], 1)\n",
" thetaBest, err = SGD(\n",
" h,\n",
" J,\n",
" dJ,\n",
" theta,\n",
" X,\n",
" Y,\n",
" alpha=1,\n",
" adaGrad=True,\n",
" maxEpochs=2500,\n",
" batchSize=100,\n",
" logError=True,\n",
" validate=0.25,\n",
" valStep=1,\n",
" lamb=0.01,\n",
" trainsetsize=m,\n",
" )\n",
" return err[1][-1], err[3][-1]\n",
"\n",
"\n",
"def plot_learning_curve():\n",
" \"\"\"Wykres krzywej uczenia się\"\"\"\n",
" plt.figure(figsize=(16, 8))\n",
" ax = plt.subplot(111)\n",
" M = np.arange(0.3, 1.0, 0.05)\n",
" Costs = [cost_trainsetsize_fun(m) for m in M]\n",
" CostTrain = [cost[0] for cost in Costs]\n",
" CostCV = [cost[1] for cost in Costs]\n",
" plt.plot(M, CostTrain, lw=3, label=\"training error\")\n",
" plt.plot(M, CostCV, lw=3, label=\"validation error\")\n",
" ax.set_xlabel(\"trainset size\")\n",
" ax.set_ylabel(\"cost\")\n",
" plt.legend()\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"### Krzywa uczenia a obciążenie i wariancja\n",
"\n",
"Wykreślenie krzywej uczenia pomaga diagnozować nadmierne i niedostateczne dopasowanie:\n",
"\n",
"<img width=\"100%\" src=\"learning-curves.png\"/>\n",
"\n",
"Źródło: http://www.ritchieng.com/machinelearning-learning-curve"
]
},
{
"cell_type": "code",
"execution_count": 33,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1600x800 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plot_learning_curve()\n"
]
}
],
"metadata": {
"celltoolbar": "Slideshow",
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.6"
},
"livereveal": {
"start_slideshow_at": "selected",
"theme": "white"
}
},
"nbformat": 4,
"nbformat_minor": 4
}