|
O‘ZBEKISTON RESPUBLIKASI AXBOROT TEXNOLOGIYALARI VA
KOMMUNIKATSIYALARINI RIVOJLANTIRISH VAZIRLIGI
MUHAMMAD AL-XORAZMIY NOMIDAGI
TOSHKENT AXBOROT TEXNOLOGIYALARI UNIVERSITETI
Kafedra Sun’iy intellekt
Fan Mashinali o‘qitishga kirish
AMALIY ISH №3
Mavzu: Mashinali o‘qitishda klassik sinflashtirish algoritmlari va ularni dasturlash
Ta’lim yo‘nalishi: Kompyuter injiniringi
Guruh: IML001-2
Bajardi: Abduhakimov Fayzullo
Tekshirdi: Ochilov Mannon
Toshkent – 2022
3-amaliy ish
Mavzu: Mashinali o‘qitishda klassik sinflashtirish algoritmlari va ularni dasturlash
№
|
Masala
|
O‘zgaruvchi tanlamadagi misollar soni
|
Sinflar soni
|
Xususiyatlari soni
|
2
|
O‘simta xavflilik darajasini bashorat qiling
|
60
|
2
|
2
|
Ishni bajarish tartibi:
Variantdagi berilgan shartni qanoatlantiruvchi datasetni shakllantiramiz. O‘simtaning xususiyatlari: uning o‘lchami va rivojlanish davri. Pythonda ushbu datasetni yaratib olamiz:
import numpy as np
#o‘simta kattaligi, rivojlanish davri
data = np.array([
[1, 10],
[1.1, 13],
[1.67, 33],
[1.2, 15],
[1.24, 20],
[1.34, 57],
[1.98, 42],
[4.23, 70],
[2.29, 35],
[2.34, 47],
[1.32, 33],
[1.4, 155],
[1.48, 27],
[1.75, 38],
[5.7, 33],
[1.8, 77],
[1.29, 21],
[2.1, 49],
[2.3, 22],
[2.39, 37],
[2.42, 56],
[2.41, 43],
[3.37, 73],
[2.2, 40],
[2.04, 38],
[2.16, 77],
[1.91, 22],
[1.78, 35],
[2.55, 43],
[3.4, 78],
[3.13, 50],
[3.12, 61],
[3.39, 89],
[4.03, 79],
[4.02, 85],
[3.34, 97],
[4.13, 92],
[4.24, 64],
[4.55, 92],
[4.98, 101],
[4.73, 69],
[4.61, 94],
[2.67, 56],
[5.1, 98],
[4.44, 89],
[4.41, 80],
[4.67, 99],
[4.77, 11],
[5.21, 103],
[5.24, 99],
[5.98, 122],
[5.88, 120],
[5.35, 132],
[5.87, 138],
[6.01, 147],
[6.03, 135],
[6.08, 109],
[5.77, 127],
[5.11, 92],
[5.14, 73],
])
target = np.array([0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1])
target_name = np.array(['Yengil shakli', 'Og\'ir shakli'])
2. Yaratilgan datasetning 2 ta xususiyatini olgan holda matplotlib kutubxonasi orqali grafik hosil qilamiz:
X = data[:, :2]
Y = target
from matplotlib import pyplot as plt
plt.figure(figsize=(6,6))
plt.scatter(*X[Y==0].T,s=30, alpha=1,label=target_name[0],color='red')
plt.scatter(*X[Y==1].T,s=30, alpha=1,label=target_name[1],color='green')
plt.xlabel('O\'simta kattaligi')
plt.ylabel('O\'simta davri')
plt.legend()
plt.show()
1-rasm. O’simta kattaligining o’simta davriga bog’liqlik grafigi
3. Yaratilgan datasetni modelni o‘qitish uchun 85 % va testlash uchun 15% nisbatda bo‘laklarga ajratamiz.
# Import Library for splitting data
from sklearn.model_selection import train_test_split
# Creating Train and Test datasets
X_train, X_test, y_train, y_test = train_test_split(X,Y, random_state = 42, test_size = 0.15)
from sklearn.preprocessing import StandardScaler
scaler = StandardScaler()
scaler.fit(X_train)
X_train = scaler.transform(X_train)
X_test = scaler.transform(X_test)
4. a) KNN sinflashtirish modelini quramiz:
#KNN
from sklearn.neighbors import KNeighborsClassifier
classifier = KNeighborsClassifier(n_neighbors =2)
classifier.fit(X_train, y_train)
Endi ushbu model asosida test va train to’plami uchun aniqliklarni topamiz:
y_pred_test = classifier.predict(X_test)
y_pred_train = classifier.predict(X_train)
from sklearn.metrics import classification_report, confusion_matrix, accuracy_score
result = confusion_matrix(y_test, y_pred_test)
print("Confusion Matrix:")
print(result)
result1 = classification_report(y_test, y_pred_test)
print("Classification Report:",)
print (result1)
result2 = accuracy_score(y_test,y_pred_test)
print("Accuracy of test:",result2)
result3 = accuracy_score(y_train,y_pred_train)
print("Accuracy of train:",result3)
Ko’rinib turibdiki, ushbu modelda test aniqligi 88% va train to’plami uchun esa 94% ni tashkil etmoqda.
b) SVM modelini quramiz:
from sklearn.svm import SVC
clf = SVC(kernel='linear')
clf.fit(X_train,y_train)
Endi ushbu model asosida test va train to’plami uchun aniqliklarni topamiz:
y_pred_testSVM = classifier.predict(X_test)
y_pred_trainSVM = classifier.predict(X_train)
resultSVM = confusion_matrix(y_test, y_pred_testSVM)
print("Confusion Matrix:")
print(resultSVM)
result1SVM = classification_report(y_test, y_pred_testSVM)
print("Classification Report:",)
print (result1SVM)
result2SVM = accuracy_score(y_test,y_pred_testSVM)
print("Accuracy of test:",result2SVM)
result3SVM = accuracy_score(y_train,y_pred_trainSVM)
print("Accuracy of train:",result3SVM)
Ushbu modelda ham xuddi KNN singari bir xil natijani oldik.
c) DT modelini quramiz:
from sklearn.tree import DecisionTreeClassifier
clf = DecisionTreeClassifier()
clf.fit(X_train,y_train)
Endi ushbu model asosida test va train to’plami uchun aniqliklarni topamiz:
y_pred_testDT = classifier.predict(X_test)
x_pred_trainDT = classifier.predict(X_train)
resultDT = confusion_matrix(y_test, y_pred_test)
print("Confusion Matrix:")
print(resultDT)
result1DT = classification_report(y_test, y_pred_test)
print("Classification Report:",)
print (result1DT)
result2DT = accuracy_score(y_test,y_pred_test)
print("Accuracy of test:",result2DT)
result3DT = accuracy_score(y_train,y_pred_train)
print("Accuracy of train:",result3DT)
d) RF modelini quramiz:
#RF
from sklearn.ensemble import RandomForestClassifier
clf = RandomForestClassifier(random_state=1)
clf.fit(X_train,y_train)
Endi ushbu model asosida test va train to’plami uchun aniqliklarni topamiz:
y_pred_test = classifier.predict(X_test)
y_pred_train = classifier.predict(X_train)
result = confusion_matrix(y_test, y_pred_test)
print("Confusion Matrix:")
print(result)
result1 = classification_report(y_test, y_pred_test)
print("Classification Report:",)
print (result1)
result2 = accuracy_score(y_test,y_pred_test)
print("Accuracy of test:",result2)
result3 = accuracy_score(y_train,y_pred_train)
print("Accuracy of train:",result3)
Har bir modelda test va train to’plamlari uchun aniqlik(accuracy) aniqlandi. Test to’plam uchun tartibsizlik matritsalari aniqlandi.
5. Ko’rib chiqilgan har bir model qiyosiy tahlilini amalga oshiramiz.
Test to’plam uchun:
plt.figure(figsize=(6,3))
plt.title('Sinflashtirish algoritmlarning solishtirma grafigi')
plt.bar(['KNN','SVM','Decision Tree','Random Forest'],[88,88,88,88],color='b')
plt.ylabel("Aniqlik")
plt.show()
2-rasm. Test to’plam uchun qurilgan sinflashtirish modellarining qiyosiy tahlili.
Train to’plam uchun:
plt.figure(figsize=(6,3))
plt.title('Sinflashtirish algoritmlarning solishtirma grafigi')
plt.bar(['KNN','SVM','Decision Tree','Random Forest'],[94,94,94,94],color='b')
plt.ylabel("Aniqlik")
plt.show()
3-rasm. Train to’plam uchun qurilgan sinflashtirish modellarining qiyosiy tahlili.
Do'stlaringiz bilan baham: |
