전문가를 위한 TensorFlow 2 빠른 시작

이것은 Google Colaboratory 노트북 파일입니다. Python 프로그램은 브라우저에서 직접 실행되므로 TensorFlow를 배우고 사용하기에 좋습니다. 이 튜토리얼을 따르려면 이 페이지 상단에 있는 버튼을 클릭하여 Google Colab에서 노트북을 실행하세요.

파이썬 런타임(runtime)에 연결하세요: 메뉴 막대의 오른쪽 상단에서 CONNECT를 선택하세요.
노트북의 모든 코드 셀(cell)을 실행하세요: Runtime > Run all을 선택하세요.

TensorFlow 2를 다운로드하여 설치합니다. TensorFlow를 프로그램으로 가져옵니다.

참고: TensorFlow 2 패키지를 설치하려면 pip를 업그레이드하세요. 자세한 내용은 설치 가이드를 참조하세요.

TensorFlow를 프로그램으로 가져옵니다.

import tensorflow as tf
print("TensorFlow version:", tf.__version__)

from tensorflow.keras.layers import Dense, Flatten, Conv2D
from tensorflow.keras import Model

2022-12-14 22:35:50.530976: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory
2022-12-14 22:35:50.531082: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory
2022-12-14 22:35:50.531092: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.
TensorFlow version: 2.11.0

MNIST 데이터셋을 로드하여 준비합니다.

mnist = tf.keras.datasets.mnist

(x_train, y_train), (x_test, y_test) = mnist.load_data()
x_train, x_test = x_train / 255.0, x_test / 255.0

# Add a channels dimension
x_train = x_train[..., tf.newaxis].astype("float32")
x_test = x_test[..., tf.newaxis].astype("float32")

tf.data를 사용하여 데이터셋을 섞고 배치를 만듭니다:

train_ds = tf.data.Dataset.from_tensor_slices(
    (x_train, y_train)).shuffle(10000).batch(32)

test_ds = tf.data.Dataset.from_tensor_slices((x_test, y_test)).batch(32)

케라스(Keras)의 모델 서브클래싱(subclassing) API를 사용하여 tf.keras 모델을 만듭니다:

class MyModel(Model):
  def __init__(self):
    super(MyModel, self).__init__()
    self.conv1 = Conv2D(32, 3, activation='relu')
    self.flatten = Flatten()
    self.d1 = Dense(128, activation='relu')
    self.d2 = Dense(10)

  def call(self, x):
    x = self.conv1(x)
    x = self.flatten(x)
    x = self.d1(x)
    return self.d2(x)

# Create an instance of the model
model = MyModel()

훈련에 필요한 옵티마이저(optimizer)와 손실 함수를 선택합니다:

loss_object = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)

optimizer = tf.keras.optimizers.Adam()

모델의 손실과 성능을 측정할 지표를 선택합니다. 에포크가 진행되는 동안 수집된 측정 지표를 바탕으로 최종 결과를 출력합니다.

train_loss = tf.keras.metrics.Mean(name='train_loss')
train_accuracy = tf.keras.metrics.SparseCategoricalAccuracy(name='train_accuracy')

test_loss = tf.keras.metrics.Mean(name='test_loss')
test_accuracy = tf.keras.metrics.SparseCategoricalAccuracy(name='test_accuracy')

tf.GradientTape를 사용하여 모델을 훈련합니다:

@tf.function
def train_step(images, labels):
  with tf.GradientTape() as tape:
    # training=True is only needed if there are layers with different
    # behavior during training versus inference (e.g. Dropout).
    predictions = model(images, training=True)
    loss = loss_object(labels, predictions)
  gradients = tape.gradient(loss, model.trainable_variables)
  optimizer.apply_gradients(zip(gradients, model.trainable_variables))

  train_loss(loss)
  train_accuracy(labels, predictions)

이제 모델을 테스트합니다:

@tf.function
def test_step(images, labels):
  # training=False is only needed if there are layers with different
  # behavior during training versus inference (e.g. Dropout).
  predictions = model(images, training=False)
  t_loss = loss_object(labels, predictions)

  test_loss(t_loss)
  test_accuracy(labels, predictions)

EPOCHS = 5

for epoch in range(EPOCHS):
  # Reset the metrics at the start of the next epoch
  train_loss.reset_states()
  train_accuracy.reset_states()
  test_loss.reset_states()
  test_accuracy.reset_states()

  for images, labels in train_ds:
    train_step(images, labels)

  for test_images, test_labels in test_ds:
    test_step(test_images, test_labels)

  print(
    f'Epoch {epoch + 1}, '
    f'Loss: {train_loss.result()}, '
    f'Accuracy: {train_accuracy.result() * 100}, '
    f'Test Loss: {test_loss.result()}, '
    f'Test Accuracy: {test_accuracy.result() * 100}'
  )

Epoch 1, Loss: 0.13726592063903809, Accuracy: 95.95333099365234, Test Loss: 0.065406933426857, Test Accuracy: 97.79999542236328
Epoch 2, Loss: 0.04497866705060005, Accuracy: 98.60166931152344, Test Loss: 0.05809991806745529, Test Accuracy: 98.22000122070312
Epoch 3, Loss: 0.02388325147330761, Accuracy: 99.24166870117188, Test Loss: 0.05342353880405426, Test Accuracy: 98.30999755859375
Epoch 4, Loss: 0.013876602053642273, Accuracy: 99.5433349609375, Test Loss: 0.06588396430015564, Test Accuracy: 98.23999786376953
Epoch 5, Loss: 0.009471668861806393, Accuracy: 99.68499755859375, Test Loss: 0.06574567407369614, Test Accuracy: 98.25999450683594

훈련된 이미지 분류기는 이 데이터셋에서 약 98%의 정확도를 달성합니다. 더 자세한 내용은 TensorFlow 튜토리얼을 참고하세요.