本帖是学习Sebastian Raschka 的《Python Machine Learning》做的笔记,便于需要时查看。
Chapter 2 Training Machine Learning Algorithms for Classification including:
- building an intuition for machine learning algorithms
- using pandas, NumPy, matplotlib to read in, process, visualize data
- implementing linear classification algprithms in Python
1. Artificial heurons - a brief glimpse into the early history of machine learning
Warren McCullock and Walter Pitts 发表si mplified brain cell 概念,称作McCullock-Pitts (MCP) neuron, in 1943 (W. S. McCulloch and W. Pitts). A Logical Calculus of the Ideas Immanent in Nervous Activity.
McCullock and Pitts described such a nerve cell as a simple logic gate with binary outputs; multiple signals arrive at the dendrites, are then integrated into the cell body, and, if the accumulated signal exceeds a certain threshold, an output signal is generated that will be passed on by the axon.
Frank Rosenblatt 在MCP神经模型的基础上发表感知机学习规则(perceptron learning rule)(F. Rosenblatt, The Perceptron, a Perceiving and Recognizing Automaton. Cornell Aeronautical Laboratory, 1957). Rosenblatt发布的算法自动学习最佳的系数,然后乘以特征值,预测新样本的类别。
二分类问题,有2个class,class=1代表positive, class=-1代表negative。定义激活函数(activation function)组合输入值x和系数w,:
,
激活函数是一种unit step function,也称为Heaviside step function:
MCP和Rosenblatt的思想是用简化论方法模拟单个神经在大脑里运转。感知机准则可以分为两步:
- 原始系数为0或者很小的随机数
- 每个训练样本计算,并更新系数。
每个输出值通过unit step function 分类,系数更新为:. 此处. 是学习率,介于(0,1),是第i个样本的真实值,是预测值。
the figure illustrates the general concept of the perceptron:
2. Implementing a perceptron learning algorithm in python
import numpy as np
class Perceptron(object):
"""Perceptron classifier.
Parameters
------------
eta : float Learning rate (between 0.0 and 1.0)
n_iter : int Passes over the training dataset.
Attributes
-----------
w_ : 1d-array Weights after fitting.
errors_ : list Number of misclassifications in every epoch.
"""
def __init__(self, eta=0.01, n_iter=10):
self.eta = eta
self.n_iter = n_iter
def fit(self, X, y):
"""Fit training data.
Parameters
----------
X : {array-like}, shape = [n_samples, n_features]
Training vectors, where n_samples is the number of samples and
n_features is the number of features.
y : array-like, shape = [n_samples]
Target values.
Returns
-------
self : object
"""
self.w_ = np.zeros(1 + X.shape[1])
self.errors_ = []
for _ in range(self.n_iter):
errors = 0
for xi, target in zip(X, y):
update = self.eta * (target - self.predict(xi))
self.w_[1:] += update * xi
self.w_[0] += update
errors += int(update != 0.0)
self.errors_.append(errors)
return self
def net_input(self, X):
"""Calculate net input"""
return np.dot(X, self.w_[1:]) + self.w_[0]
def predict(self, X):
"""Return class label after unit step"""
return np.where(self.net_input(X) >= 0.0, 1, -1)
3. Training a perceptron model on the Iris dataset
Iris包含150个iris flowers样本,分三个品种:Setosa, Versicolor and Viriginica. 特征有Sepal length, Sepal width, Petal length, Petal width, class label. 下面我们只看 Setosa和Versicolor两个品种的sepal length 和petal length的data。
先从pandas库中加载数据,查看数据的加载情况:
import pandas as pd
df = pd.read_csv('https://archive.ics.uci.edu/ml/'
'machine-learning-databases/iris/iris.data', header=None)
df.tail()
再根据要求提取数据,提取品种为Setosa和Versicolor,特征为sepal length和petal length的数据,将品种为Versicolor的class标为1,品种为Setosa的class标为-1,将其可视化:
import matplotlib.pyplot as plt
y = df.iloc[0:100, 4].values
y = np.where(y == 'Iris-setosa', -1, 1)
X = df.iloc[0:100, [0, 2]].values
plt.scatter(X[:50, 0], X[:50, 1],
color='red', marker='o', label='setosa')
plt.scatter(X[50:100, 0], X[50:100, 1],
color='blue', marker='x', label='versicolor')
plt.xlabel('petal length')
plt.ylabel('sepal length')
plt.legend(loc='upper left')
plt.show()
接着用感知机算法训练iris数据,并绘制出每次迭代误分的个数:
ppn = Perceptron(eta=0.1, n_iter=10)
ppn.fit(X, y)
plt.plot(range(1, len(ppn.errors_) + 1), ppn.errors_, marker='o')
plt.xlabel('Epochs')
plt.ylabel('Number of misclassifications')
plt.show()
从图中可以看出,迭代到第6次时,训练集的分类已经完美了,下面可视化判定边界:
from matplotlib.colors import ListedColormap
def plot_decision_regions(X, y, classifier, resolution=0.02):
# setup marker generator and color map
markers = ('s', 'x', 'o', '^', 'v')
colors = ('red', 'blue', 'lightgreen', 'gray', 'cyan')
cmap = ListedColormap(colors[:len(np.unique(y))])
# plot the decision surface
x1_min, x1_max = X[:, 0].min() - 1, X[:, 0].max() + 1
x2_min, x2_max = X[:, 1].min() - 1, X[:, 1].max() + 1
xx1, xx2 = np.meshgrid(np.arange(x1_min, x1_max, resolution),
np.arange(x2_min, x2_max, resolution))
Z = classifier.predict(np.array([xx1.ravel(), xx2.ravel()]).T)
Z = Z.reshape(xx1.shape)
plt.contourf(xx1, xx2, Z, alpha=0.4, cmap=cmap)
plt.xlim(xx1.min(), xx1.max())
plt.ylim(xx2.min(), xx2.max())
# plot class samples
for idx, cl in enumerate(np.unique(y)):
plt.scatter(x=X[y == cl, 0], y=X[y == cl, 1], alpha=0.8, c=cmap(idx),
marker=markers[idx], label=cl)
plot_decision_regions(X, y, classifier=ppn)
plt.xlabel('sepal length [cm]')
plt.ylabel('petal length [cm]')
plt.legend(loc='upper left')
plt.show()
4. Adaptive linear neurons and the convergence of learning
ADAptive LInear NEuron (Adaline)是Bernard Widrow提出的,Adaline提出了最小化损失函数(minimizing cost function),系数更新基于linear activation function而不是unit step function。过程如下图:
上图中,用linear activation function计算model error和 update weights,而不是二分类标签。
5. Minimizing cost functions with gradient decent
监督学习中,定义目标函数(objective function)优化训练过程是一个重要的组成部分。目标函数也就是损失函数要去最小化的。在Adaline中,用cost function J学习weights sum of squared errors(SSE):
利用梯度下降(gradient descent)找最小化cost function 的weights。梯度下降是下山(climbing down a hill)直到局部或全局损失最小。在每次迭代中,step size由学习率(learning rate)和梯度的斜率(slope)决定。
update the weights by taking a step away from the gradient of cost function :
the weigh change is defined as the negative gradient multiplied by learning rate :
to compute the gradient of the cost function, we need to compute the partial derivative of cost function with respect to each weight : , therefore, .
6. Implementing an Adaline Linear Neuron in Python
class AdalineGD(object):
"""ADAptive LInear NEuron classifier.
Parameters
------------
eta : float Learning rate (between 0.0 and 1.0)
n_iter : int
Passes over the training dataset.
Attributes
-----------
w_ : 1d-array
Weights after fitting.
errors_ : list Number of misclassifications in every epoch.
"""
def __init__(self, eta=0.01, n_iter=50):
self.eta = eta
self.n_iter = n_iter
def fit(self, X, y):
""" Fit training data.
Parameters
----------
X : {array-like}, shape = [n_samples, n_features]
Training vectors,
where n_samples is the number of samples and
n_features is the number of features.
y : array-like, shape = [n_samples]
Target values.
Returns
-------
self : object
"""
self.w_ = np.zeros(1 + X.shape[1])
self.cost_ = []
for i in range(self.n_iter):
output = self.net_input(X)
errors = (y - output)
self.w_[1:] += self.eta * X.T.dot(errors)
self.w_[0] += self.eta * errors.sum()
cost = (errors**2).sum() / 2.0
self.cost_.append(cost)
return self
def net_input(self, X):
"""Calculate net input"""
return np.dot(X, self.w_[1:]) + self.w_[0]
def activation(self, X):
"""Compute linear activation"""
return self.net_input(X)
def predict(self, X):
"""Return class label after unit step"""
return np.where(self.activation(X) >= 0.0, 1, -1)
对比不同学习率的结果:
fig, ax = plt.subplots(nrows=1, ncols=2, figsize=(8, 4))
ada1 = AdalineGD(n_iter=10, eta=0.01).fit(X, y)
ax[0].plot(range(1, len(ada1.cost_) + 1), np.log10(ada1.cost_), marker='o')
ax[0].set_xlabel('Epochs')
ax[0].set_ylabel('log(Sum-squared-error)')
ax[0].set_title('Adaline - Learning rate 0.01')
ada2 = AdalineGD(n_iter=10, eta=0.0001).fit(X, y)
ax[1].plot(range(1, len(ada2.cost_) + 1), ada2.cost_, marker='o')
ax[1].set_xlabel('Epochs')
ax[1].set_ylabel('Sum-squared-error')
ax[1].set_title('Adaline - Learning rate 0.0001')
plt.show()
当学习率很小时,就要求很多次迭代去收敛。
梯度下降需要特征缩放(feature scaling),这里用standardization,即将数据标准正态分布化。先将数据标准化,再用Adaline
X_std = np.copy(X)
X_std[:,0] = (X[:,0] - X[:,0].mean()) / X[:,0].std()
X_std[:,1] = (X[:,1] - X[:,1].mean()) / X[:,1].std()
ada = AdalineGD(n_iter=15, eta=0.01)
ada.fit(X_std, y)
plot_decision_regions(X_std, y, classifier=ada)
plt.title('Adaline - Gradient Descent')
plt.xlabel('sepal length [standardized]')
plt.ylabel('petal length [standardized]')
plt.legend(loc='upper left')
plt.show()
plt.plot(range(1, len(ada.cost_) + 1), ada.cost_, marker='o')
plt.xlabel('Epochs')
plt.ylabel('Sum-squared-error')
plt.show()
7. Large scale machine learning and stochastic gradient descent
用stochastic gradient descent (也称为iterative or on-line gradient descent)解决大规模问题.
update the weights incrementally for each training sample:
随机梯度下降能更快的收敛,因为系数更新更频繁了。由于每个梯度基于单个训练样本计算,error surface is noisier。此外,随机梯度下降可以用于online learning。
python code 如下:
from numpy.random import seed
class AdalineSGD(object):
"""ADAptive LInear NEuron classifier.
Parameters
------------
eta : float
Learning rate (between 0.0 and 1.0)
n_iter : int
Passes over the training dataset.
Attributes
-----------
w_ : 1d-array
Weights after fitting.
errors_ : list
Number of misclassifications in every epoch.
shuffle : bool (default: True)
Shuffles training data every epoch
if True to prevent cycles.
random_state : int (default: None)
Set random state for shuffling
and initializing the weights.
"""
def __init__(self, eta=0.01, n_iter=10, shuffle=True, random_state=None):
self.eta = eta
self.n_iter = n_iter
self.w_initialized = False
self.shuffle = shuffle
if random_state:
seed(random_state)
def fit(self, X, y):
""" Fit training data.
Parameters
----------
X : {array-like}, shape = [n_samples, n_features]
Training vectors, where n_samples is the number of samples and
n_features is the number of features.
y : array-like, shape = [n_samples]
Target values.
Returns
-------
self : object
"""
self._initialize_weights(X.shape[1])
self.cost_ = []
for i in range(self.n_iter):
if self.shuffle:
X, y = self._shuffle(X, y)
cost = []
for xi, target in zip(X, y):
cost.append(self._update_weights(xi, target))
avg_cost = sum(cost)/len(y)
self.cost_.append(avg_cost)
return self
def partial_fit(self, X, y):
"""Fit training data without reinitializing the weights"""
if not self.w_initialized:
self._initialize_weights(X.shape[1])
if y.ravel().shape[0] > 1:
for xi, target in zip(X, y):
self._update_weights(xi, target)
else:
self._update_weights(X, y)
return self
def _shuffle(self, X, y):
"""Shuffle training data"""
r = np.random.permutation(len(y))
return X[r], y[r]
def _initialize_weights(self, m):
"""Initialize weights to zeros"""
self.w_ = np.zeros(1 + m)
self.w_initialized = True
def _update_weights(self, xi, target):
"""Apply Adaline learning rule to update the weights"""
output = self.net_input(xi)
error = (target - output)
self.w_[1:] += self.eta * xi.dot(error)
self.w_[0] += self.eta * error
cost = 0.5 * error**2
return cost
def net_input(self, X):
"""Calculate net input"""
return np.dot(X, self.w_[1:]) + self.w_[0]
def activation(self, X):
"""Compute linear activation"""
return self.net_input(X)
def predict(self, X):
"""Return class label after unit step"""
return np.where(self.activation(X) >= 0.0, 1, -1)
_shuffle: via the permutation function in np.random, generate a random sequence of unique numbers in the range 0 to 100, these numbers used as indices to shuffle feature matrix and class label vector.
plot the training results:
ada = AdalineSGD(n_iter=15, eta=0.01, random_state=1)
ada.fit(X_std, y)
plot_decision_regions(X_std, y, classifier=ada)
plt.title('Adaline - Stochastic Gradient Descent')
plt.xlabel('sepal length [standardized]')
plt.ylabel('petal length [standardized]')
plt.legend(loc='upper left')
plt.show()
plt.plot(range(1, len(ada.cost_) + 1), ada.cost_, marker='o')
plt.xlabel('Epochs')
plt.ylabel('Average Cost')
plt.show()