|
Konteynerlashtirishdan foydalanish
|
| bet | 13/182 | | Sana | 19.05.2024 | | Hajmi | 5,69 Mb. | | #244351 |
Bog'liq Python sun\'iy intellekt texnologiyasi Dasrlik 2024Konteynerlashtirishdan foydalanish:
Docker kabi konteynerlar AI modellarini turli muhitlarga joylashtirishni osonlashtirishi va ularning izolyatsiyasini ta’minlashi mumkin.
Modelning hayot aylanishini boshqarishning avtomatlashtirilgan tizimlarini joriy etish:
Modellarni ishlab chiqish, o‘qitish va joylashtirishni osonlashtirish uchun MLflow yoki Kubeflow kabi model hayot aylanishini boshqarish platformalaridan foydalanish.
Python ilovalariga integratsiyalashda xavfsizlik, ishlash va miqyoslilik talablarini hisobga olish muhimdir. Tegishli texnologiyalar va kutubxonalarni tanlash sizning loyihangizning aniq vazifalari va ehtiyojlariga bog’liq.
Python-da scikit-learn, TensorFlow va PyTorch-dan foydalangan holda mashinali o‘qitish kutubxonalaridan foydalanishning oddiy misollarini keltiraman.
Scikit-learn yordamida misol:
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score
# Ma’lumotlarni yuklash (Iris ma’lumotlar to‘plamiga misol)
from sklearn.datasets import load_iris
iris = load_iris()
X, y = iris.data, iris.target
# Ma’lumotlarni o‘qitish va test namunalariga bo‘lish
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
# Modelni yaratish (tasodifiy o‘rmon klassifikatori)
model = RandomForestClassifier(n_estimators=100, random_state=42)
# Modelni o‘qitish
model.fit(X_train, y_train)
# Sinov namunasidagi bashorat
predictions = model.predict(X_test)
# Modelning aniqligini baholash
accuracy = accuracy_score(y_test, predictions)
print(f"Accuracy: {accuracy}")
2. Tens yoki Flow yordamida misol (oddiy neyron tarmoq uchun):
import tensorflow as tf
from tensorflow.keras import layers, models
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
# Ma’lumotlarni yuklash (Iris ma’lumotlar to‘plamiga misol)
from sklearn.datasets import load_iris
iris = load_iris()
X, y = iris.data, iris.target
# Ma’lumotlarni o‘qitish va test namunalariga bo‘lish
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
# Oddiy neyron tarmog’ini qurish
model = models.Sequential([
layers.Dense(64, activation='relu', input_shape=(4,)),
layers.Dense(3, activation='softmax') ])
# Modelni kompilyatsiya qilish
model.compile(optimizer='adam',
loss='sparse_categorical_crossentropy',
metrics=['accuracy'])
# Modelni o‘qitish
model.fit(X_train, y_train, epochs=50, batch_size=32, validation_split=0.2)
# Modelning aniqligini baholash
test_loss, test_acc = model.evaluate(X_test, y_test)
print(f"Test Accuracy: {test_acc}")
|
| |
