MUHAMMAD AL-XORAZMIY NOMIDAGI TOSHKENT AXBOROT TEXNOLOGIYALARI UNIVERSITETI
“KIBERXAVFSIZLIK” FAKULTETI
“Mashinali oqitish kirish” fanidan tayyorlagan
2-AMALIY ISHI
Topshirdi: Madatov I.
Tekshirdi: Ochilov M.
Guruh: 713-21
TOSHKENT - 2023
2-amaliy ish.
6-variant
Berilgan variantdagi masala yuzasidan o’rgatuvchi tanlama(dataset)ni
shakllantiring.
dataset=np.array([#yosh, ma'lumot, tajriba, investitsiyalar
[30,5,2,1,1],
[25,3,1,3,0],
[35,10,4,6,1],
[28,6,3,9,0],
[40,8,2,3,1],
[22,2,1,1,0],
[32,7,3,1,1],
[27,4,2,0,0],
[29,6,2,3,1],
[24,3,1,0,0]
])
.
Birinchi to'rtta qiymat xususiyatlarni ifodalaydi (yosh, ma'lumot, tajriba, investitsiyalar).
2.Yaratilgan dataset ning ixtiyoriy ikkita xususiyatini olgan holda matplotlib kutubxonasidan foydalanib grafik tasvirlang.
x = dataset[:,1]#malumot
y = dataset[:,2]#tajriba
c = dataset[:,-1]#sinf
plt.figure(figsize=(6,4))
plt.scatter(x[c==0], y[c==0], s=30, alpha=1, label='1-sinf', color='r', marker='s')
plt.scatter(x[c==1], y[c==1], s=30, alpha=0.8, label='2-sinf', color='b', marker='^')
plt.xlabel('Ma\'lumot')
plt.ylabel('Tajriba')
plt.legend()
plt.grid()
plt.show()
Datasetni malumoti va tajribasi xususiyati bo’yicha grafik chizish kodi.
1-rasm. Datasetni matplotlib kutubxonasi yordamida ikkita xususiyatininng grafigi.
3.Yaratilgan datasetni modelni o’qitish uchun 85 % va testlash uchun 15% nisbatda bo’laklarga ajrating.
rasm. Datasetda model qurish.
X_train=dataset[:,:-1]
Y_train=dataset[:,-1]
x_train,x_test,y_train,y_test= train_test_split(X_train,Y_train,test_size=0.15,random_state=42)
3-Sklearn kutubxonasidan foydalangan holda logistik_regressiya modelini quring
#LogisticRegression
logisticRegr = LogisticRegression()
logisticRegr.fit(x_train,y_train)
#for train
train_pred=logisticRegr.predict(x_train)
4-Model aniqligini hisoblang(o’rgatuvchi tanalama uchun).
score = logisticRegr.score(x_train,y_train)
print(score)
5-Modelini test to’plamdagi aniqligini hisoblang
#for test
test_pred=logisticRegr.predict(x_test)
score=logisticRegr.score(x_test,y_test)
print(score)
6-Test to’plam uchun tartibsizlik matritsasi (confusion_matrix) ni hisoblang va tariflang
#confusion matrix
#for train
cm=confusion_matrix(y_train,train_pred)
print(cm)
#for tesst
cm=confusion_matrix(y_test,test_pred)
print(cm)
Arraydao‘qituvchi xuxusiyatlari berilsa uni qaysi sinfga tegishli ekanligini chiqarib beradi.
print("Logistic Regression tested : ",logisticRegr.predict([[41,6,4,2]])[0])
Natijalar:
Xulosa:
Men Logistic Regressianing mohiyatini tushindim, ya’ni sigmoyid functiondan foydalanilgan holda model tuzib olish hamda aniqlikni tekshirish qanday bolib borishi . Confusion matrixda -bu tasniflash algoritmining ishlashini aniqlash uchun ishlatiladigan jadval
|
