def predict_preferred_language_by_city(k_values, cities):
for k in k_values:
count = 0
for city in cities:
new = cities.copy()
new.remove(city)
lan = knn_classify(k, new, city[0])
if(lan == city[1]):
count+=1
print(k, "neighbor[s]:", count, "correct out of", len(cities))
"""
def predict_preferred_language_by_city(k_values, cities):
for i in k_values:
count = 0
for city in cities:
temp = cities.copy()
temp.remove(city)
if (knn_classify(i, temp, tuple(city[0])) == city[1]):
count += 1
print(i, "neighbor[s]:", count, "correct out of", len(cities))
"""
def predict_preferred_language_by_city(k_values, cities):
"""
TODO
predicts a preferred programming language for each city using above knn_classify() and
counts if predicted language matches the actual language.
Finally, print number of correct for each k value using this:
print(k, "neighbor[s]:", num_correct, "correct out of", len(cities))
"""
for k in k_values:
num_correct = 0
for loc, lang in cities:
labeled = cities.copy()
labeled.remove((loc, lang))
pred = knn_classify(k, labeled, loc)
if (pred == lang):
num_correct += 1
print(k, "neighbor[s]:", num_correct, "correct out of", len(cities))
def predict_preferred_language_by_city_scikit(k_values, cities):
print("Output after using KNN by scikit")
for k in k_values:
num_correct = 0
knn = KNeighborsClassifier(n_neighbors=k)
for city in cities:
new_list = cities.copy()
new_list.remove(city)
i, label = get_i_label(new_list)
knn.fit(i, label)
predicted_language = knn.predict(
np.array([[city[0][0], city[0][1]]]))
if predicted_language == city[1]:
num_correct += 1
print(k, "neighbor[s]:", num_correct, "correct out of", len(cities))
def predict_preferred_language_by_city(k_values, cities):
"""
TODO
predicts a preferred programming language for each city using above knn_classify() and
counts if predicted language matches the actual language.
Finally, print number of correct for each k value using this:
print(k, "neighbor[s]:", num_correct, "correct out of", len(cities))
"""
for k in k_values:
count = 0
for city in cities:
new_cities = cities.copy()
new_cities.remove(city)
new_language = knn_classify(k, new_cities, tuple(city[0]))
if (new_language == city[1]):
count += 1
print(k, "neighbor[s]:", count, "correct out of", len(cities))
def predict_preferred_language_by_city(k_values, cities):
"""
TODO
predicts a preferred programming language for each city using above knn_classify() and
counts if predicted language matches the actual language.
Finally, print number of correct for each k value using this:
print(k, "neighbor[s]:", num_correct, "correct out of", len(cities))
"""
for i in k_values:
num_correct = 0
for j, z in enumerate(cities):
new_list = cities.copy()
new_list.remove(z)
predicted_language = knn_classify(i, new_list, z[0])
if (predicted_language == cities[j][1]):
num_correct = num_correct + 1
print(i, "neighbor[s]:", num_correct, "correct out of", len(cities))
def predict_preferred_language_by_city(k_values, cities):
"""
TODO
predicts a preferred programming language for each city using above knn_classify() and
counts if predicted language matches the actual language.
Finally, print number of correct for each k value using this:
print(k, "neighbor[s]:", num_correct, "correct out of", len(cities))
"""
#knn = KNeighborsClassifier(k_values)
#knn.fit()
#l = len(cities)
for k in k_values:
num_correct = 0
for i in range(0, len(cities)):
new_pts = cities.copy()
del new_pts[i]
output = knn_classify(k, new_pts, cities[i][0])
if (output == cities[i][1]):
num_correct += 1
print(k, "neighbor[s]:", num_correct, "correct out of", len(cities))