fruit_prediction = knn.predict([[20, 4.3, 5.5]])
lookup_fruit_name[fruit_prediction[0]]
# In[12]:
# second example: a larger, elongated fruit with mass 100g, width 6.3 cm, height 8.5 cm
fruit_prediction = knn.predict([[100, 6.3, 8.5]])
lookup_fruit_name[fruit_prediction[0]]
# ### Plot the decision boundaries of the k-NN classifier
# In[13]:
from adspy_shared_utilities import plot_fruit_knn
plot_fruit_knn(X_train, y_train, 5, 'uniform') # we choose 5 nearest neighbors
# ### How sensitive is k-NN classification accuracy to the choice of the 'k' parameter?
# In[14]:
k_range = range(1, 20)
scores = []
for k in k_range:
knn = KNeighborsClassifier(n_neighbors=k)
knn.fit(X_train, y_train)
scores.append(knn.score(X_test, y_test))
plt.figure()
plt.xlabel('k')
fig = plt.figure()
ax = fig.add_subplot(111, projection='3d')
ax.scatter(X_train['width'],
X_train['height'],
X_train['color_score'],
c=y_train,
marker='o',
s=100)
ax.set_xlabel('width')
ax.set_ylabel('height')
ax.set_zlabel('color_score')
plt.show()
X = fruits[['mass', 'width', 'height']]
y = fruits['fruit_label']
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)
from sklearn.neighbors import KNeighborsClassifier
knn = KNeighborsClassifier(n_neighbors=5)
knn.fit(X_train, y_train)
knn.score(X_test, y_test)
fruit_prediction = knn.predict([[20, 4.3, 5.5]])
look_up_fruit_name[fruit_prediction[0]]
fruit_prediction = knn.predict([[100, 6.3, 8.5]])
look_up_fruit_name[fruit_prediction[0]]
from adspy_shared_utilities import plot_fruit_knn
plot_fruit_knn(X_train, y_train, 5, 'uniform')
# Use trained k-NN classifier model to classify new objects.
lookup = dict(zip(fruits.fruit_label.unique(),
fruits.fruit_name.unique())) # Label #-name pair
fruit_prediction = knn.predict(
[[20, 4.3, 5.5,
0.45]]) # Predict function, need all 4 parameters to generate.
print(lookup[fruit_prediction[0]])
fruit_prediction = knn.predict([[100, 10.2, 12.3, 0.63]])
print(lookup[fruit_prediction[0]])
# Sensitivity of k-NN classifier accuracy to choice of 'k' parameter
k_range = range(1, 20)
scores = []
for k in k_range: # k=1 to k=19
knn = KNeighborsClassifier(n_neighbors=k)
knn.fit(X_train, y_train)
scores.append(knn.score(X_test, y_test))
plt.figure()
plt.scatter(k_range, scores)
plt.xlabel('k')
plt.ylabel("accuracy")
plt.xticks([0, 5, 10, 15, 20])
# Graph displaying decision boundaries.
from adspy_shared_utilities import plot_fruit_knn
plot_fruit_knn(X_train, y_train, 5, "uniform")
fruit_prediction = knn.predict([[20, 4.3, 5.5]])
lookup_fruit_name[fruit_prediction[0]]
# In[25]:
# second example: a larger, elongated fruit with mass 100g, width 6.3 cm, height 8.5 cm
fruit_prediction = knn.predict([[100, 6.3, 8.5]])
lookup_fruit_name[fruit_prediction[0]]
# ### Plot the decision boundaries of the k-NN classifier
# In[27]:
from adspy_shared_utilities import plot_fruit_knn
plot_fruit_knn(X_train, y_train, 3,
'distance') # we choose 5 nearest neighbors
# ### How sensitive is k-NN classification accuracy to the choice of the 'k' parameter?
# In[28]:
k_range = range(1, 20)
scores = []
for k in k_range:
knn = KNeighborsClassifier(n_neighbors=k)
knn.fit(X_train, y_train)
scores.append(knn.score(X_test, y_test))
plt.figure()
plt.xlabel('k')
# train the classifier (fit the estimator) using training data
print(knn.fit(X_train, y_train))
# Estimate the accuracy of the classifier on future data using the test data
print(knn.score(X_test, y_test))
# Use the trained k-NN classifier model to classify new, previously unseen objects
fruit_prediction = knn.predict([[20, 4.3, 5.5]]) # mass, width and height
print(lookup_fruit_name[fruit_prediction[0]])
fruit_prediction = knn.predict([[100, 6.3, 8.5]]) # mass, width and height
print(lookup_fruit_name[fruit_prediction[0]])
# Plot decision boundaries of the k-NN classifier
from adspy_shared_utilities import plot_fruit_knn
plot_fruit_knn(X_train, y_train, 5, 'uniform')
plot_fruit_knn(X_train, y_train, 5, 'distance')
# How sensitive is k-NN classification accuracy to the choice of the 'k' parameter?
k_range = range(1, 20)
scores = []
for k in k_range:
knn = KNeighborsClassifier(n_neighbors=k)
knn.fit(X_train, y_train)
scores.append(knn.score(X_test, y_test))
plt.figure()
plt.xlabel('k')
plt.ylabel('accuracy')
plt.scatter(k_range, scores)
def decisionBoundaries():
from adspy_shared_utilities import plot_fruit_knn
plot_fruit_knn(X_train, y_train, 5,
'uniform') # we choose 5 nearest neighbors
