KNN klasifikacija¶

Biblioteke¶

In [1]:
import pandas as pd

Standard scales - Klasa koja transofrmise numericke atriute tako da budu iz normalne raspodele

Podaci¶

In [2]:
df = pd.read_csv("../data/iris.csv")
In [3]:
df.head()
Out[3]:
Sepal_Length Sepal_Width Petal_Length Petal_Width Species
0 5.1 3.5 1.4 0.2 setosa
1 4.9 3.0 1.4 0.2 setosa
2 4.7 3.2 1.3 0.2 setosa
3 4.6 3.1 1.5 0.2 setosa
4 5.0 3.6 1.4 0.2 setosa
In [4]:
df.describe()
Out[4]:
Sepal_Length Sepal_Width Petal_Length Petal_Width
count 150.000000 150.000000 150.000000 150.000000
mean 5.843333 3.057333 3.758000 1.199333
std 0.828066 0.435866 1.765298 0.762238
min 4.300000 2.000000 1.000000 0.100000
25% 5.100000 2.800000 1.600000 0.300000
50% 5.800000 3.000000 4.350000 1.300000
75% 6.400000 3.300000 5.100000 1.800000
max 7.900000 4.400000 6.900000 2.500000
In [5]:
features = df.columns[:-1]
features
Out[5]:
Index(['Sepal_Length', 'Sepal_Width', 'Petal_Length', 'Petal_Width'], dtype='object')
In [6]:
X = df[features]
y = df["Species"]
print(X.shape)
print(y.shape)

(150, 4)
(150,)

Preprocesiranje¶

In [7]:
from sklearn.model_selection import train_test_split
In [8]:
from sklearn.preprocessing import StandardScaler
In [9]:
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=89, stratify=y)
In [10]:
print(X_train.shape)
print(X_test.shape)
print(y_train.shape)
print(y_test.shape)

(105, 4)
(45, 4)
(105,)
(45,)
In [11]:
scaler = StandardScaler()
scaler.fit(X_train)
Out[11]:
StandardScaler()
In [12]:
print(scaler.mean_)
print(scaler.var_)
print(scaler.scale_) # Standardna decijacija

[5.83619048 3.02952381 3.76190476 1.19238095]
[0.71583311 0.18855692 3.20692971 0.58146576]
[0.84606921 0.43423141 1.79079025 0.76253902]
In [13]:
X_train = scaler.transform(X_train)
X_test = scaler.transform(X_test)

Treniranje modela¶

In [14]:
from sklearn.neighbors import KNeighborsClassifier
In [15]:
model = KNeighborsClassifier(n_neighbors=4)
model.fit(X_train, y_train)
Out[15]:
KNeighborsClassifier(n_neighbors=4)
In [16]:
classes = model.classes_

Performanse modela nad skupom za treniranje¶

In [17]:
from sklearn.metrics import confusion_matrix
In [18]:
y_train_pred = model.predict(X_train)
pd.DataFrame(confusion_matrix(y_train, y_train_pred), columns=classes, index=classes)
Out[18]:
setosa versicolor virginica
setosa 35 0 0
versicolor 0 33 2
virginica 0 3 32

Performanse modela nad skupom za testiranje¶

In [19]:
y_test_pred = model.predict(X_test)
pd.DataFrame(confusion_matrix(y_test, y_test_pred), columns=classes, index=classes)
Out[19]:
setosa versicolor virginica
setosa 15 0 0
versicolor 0 15 0
virginica 0 2 13