Based on Kevin Markham's video series: Introduction to machine learning with scikit-learn
jupyter notebook 05_model_evaluation_ta.ipynb
In [1]:
# read in the iris data
from sklearn.datasets import load_iris
iris = load_iris()
# create X (features) and y (response)
X = iris.data
y = iris.target
In [2]:
# import the class
from sklearn.linear_model import LogisticRegression
# instantiate the model (using the default parameters)
logreg = LogisticRegression()
# fit the model with data
logreg.fit(X, y)
# predict the response values for the observations in X
y_pred = logreg.predict(X)
print(y_pred)
print("{0} predictions".format(len(y_pred)))
Classification accuracy:
In [3]:
# compute classification accuracy for the logistic regression model
from sklearn import metrics
print metrics.accuracy_score(y, y_pred)
Mean Absolute Error (MAE) is the mean of the absolute value of the errors:
$$\frac 1n\sum_{i=1}^n|y_i-\hat{y}_i|$$Mean Squared Error (MSE) is the mean of the squared errors:
$$\frac 1n\sum_{i=1}^n(y_i-\hat{y}_i)^2$$Root Mean Squared Error (RMSE) is the square root of the mean of the squared errors:
$$\sqrt{\frac 1n\sum_{i=1}^n(y_i-\hat{y}_i)^2}$$http://scikit-learn.org/stable/modules/model_evaluation.html#scoring-parameter
Look back at 04_model_training
and see how high an accuracy you can achieve for different values of n_neighbors
. Try to understand why different values do better than others in terms of the pictures we saw in 04_model_training
.
You can change feature1
and feature2
in the cell below to visualize different projections of the data.
In [4]:
feature1 = 1 # feature on x axis
feature2 = 3 # feature on y axis
data = X
f1vals = X[:,feature1]
f2vals = X[:,feature2]
import numpy as np
targets = dict(zip(range(3), iris.target_names))
features = dict(zip(range(4), iris.feature_names))
%matplotlib inline
import matplotlib.pyplot as plt
colors = ['g', 'r', 'b']
fig = plt.figure(figsize=(8,8))
ax = plt.subplot()
for species in targets.keys():
f1 = f1vals[np.where(y==species)]
f2 = f2vals[np.where(y==species)]
ax.scatter(f1, f2, c=colors[species], label=targets[species], s=40)
ax.set(xlabel=features[feature1], ylabel=features[feature2])
ax.legend()
In [5]:
from sklearn.neighbors import KNeighborsClassifier
knn = KNeighborsClassifier(n_neighbors=5)
knn.fit(X, y)
y_pred = knn.predict(X)
print metrics.accuracy_score(y, y_pred)
In [6]:
knn = KNeighborsClassifier(n_neighbors=1)
knn.fit(X, y)
y_pred = knn.predict(X)
print metrics.accuracy_score(y, y_pred)
jupyter notebook 05_model_evaluation_tts.ipynb
Image Credit: Overfitting by Chabacano. Licensed under GFDL via Wikimedia Commons.
In [7]:
# print the shapes of X and y
print X.shape
print y.shape
In [8]:
# STEP 1: split X and y into training and testing sets
from sklearn.cross_validation import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4, random_state=4)
What did this accomplish?
In [9]:
# STEP 1: split X and y into training and testing sets
from sklearn.cross_validation import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4, random_state=4)
In [10]:
# print the shapes of the new X objects
print X_train.shape
print X_test.shape
In [11]:
# print the shapes of the new y objects
print y_train.shape
print y_test.shape
In [12]:
# STEP 2: train the model on the training set
logreg = LogisticRegression()
logreg.fit(X_train, y_train)
Out[12]:
In [13]:
# STEP 3: make predictions on the testing set
y_pred = logreg.predict(X_test)
# compare actual response values (y_test) with predicted response values (y_pred)
print metrics.accuracy_score(y_test, y_pred)
Repeat for KNN with K=5:
In [14]:
knn = KNeighborsClassifier(n_neighbors=5)
knn.fit(X_train, y_train)
y_pred = knn.predict(X_test)
print metrics.accuracy_score(y_test, y_pred)
Repeat for KNN with K=1:
In [15]:
knn = KNeighborsClassifier(n_neighbors=1)
knn.fit(X_train, y_train)
y_pred = knn.predict(X_test)
print metrics.accuracy_score(y_test, y_pred)
In [ ]:
# try K=1 through K=25 and record testing accuracy
k_range = range(1, 26)
scores = [] # calculate accuracies for each value of K!
#Now we plot:
import matplotlib.pyplot as plt
# allow plots to appear within the notebook
%matplotlib inline
plt.plot(k_range, scores)
plt.xlabel('Value of K for KNN')
plt.ylabel('Testing Accuracy')
In [16]:
# try K=1 through K=25 and record testing accuracy
k_range = range(1, 26)
scores = []
for k in k_range:
knn = KNeighborsClassifier(n_neighbors=k)
knn.fit(X_train, y_train)
y_pred = knn.predict(X_test)
scores.append(metrics.accuracy_score(y_test, y_pred))
In [17]:
# import Matplotlib (scientific plotting library)
import matplotlib.pyplot as plt
# allow plots to appear within the notebook
%matplotlib inline
# plot the relationship between K and testing accuracy
plt.plot(k_range, scores)
plt.xlabel('Value of K for KNN')
plt.ylabel('Testing Accuracy')
Out[17]:
In [18]:
# instantiate the model with the best known parameters
knn = KNeighborsClassifier(n_neighbors=11)
# train the model with X and y (not X_train and y_train)
knn.fit(X, y)
# make a prediction for an out-of-sample observation
print(knn.predict([[3, 5, 4, 2]]))
In [19]:
# import Matplotlib (scientific plotting library)
import matplotlib.pyplot as plt
# allow plots to appear within the notebook
%matplotlib inline
# plot the relationship between K and testing accuracy
plt.plot(k_range, scores)
plt.xlabel('Value of K for KNN')
plt.ylabel('Testing Accuracy')
Out[19]:
In [ ]: