python
import tensorflow as tf

# 創建張量
x = tf.constant([[1, 2], [3, 4]])

# 重塑張量
y = tf.reshape(x, [1, 4])

# 切片張量
z = x[:, 1]

# 連接張量
w = tf.concat([x, x], axis=0)

python
import tensorflow as tf
from tensorflow.keras import layers

# 定義模型
model = tf.keras.Sequential([
  layers.Dense(64, activation="relu", input_shape=(784,)),
  layers.Dense(10, activation="softmax")
])

# 編譯模型
model.compile(optimizer=tf.keras.optimizers.Adam(0.001),
              loss="categorical_crossentropy",
              metrics=["accuracy"])

# 訓練模型
model.fit(x_train, y_train, epochs=10)

python
import tensorflow as tf

# 定義模型
model = tf.keras.Sequential([
  tf.keras.layers.Dense(64, activation="relu"),
  tf.keras.layers.Dense(10, activation="softmax")
])

# 定義優化器和損失函數
optimizer = tf.keras.optimizers.Adam(0.001)
loss_fn = tf.keras.losses.CategoricalCrossentropy()

# 自定義訓練循環
for epoch in range(10):
  for x, y in train_dataset:
    with tf.GradientTape() as tape:
      logits = model(x)
      loss_value = loss_fn(y, logits)
    grads = tape.gradient(loss_value, model.trainable_weights)
    optimizer.apply_gradients(zip(grads, model.trainable_weights))

python
import tensorflow as tf

# 定義模型
model = tf.keras.Sequential([
  tf.keras.layers.Dense(64, activation="relu"),
  tf.keras.layers.Dense(10, activation="softmax")
])

# 定義優化器和損失函數
optimizer = tf.keras.optimizers.Adam(0.001)
loss_fn = tf.keras.losses.CategoricalCrossentropy()

# 定義分布式策略
strategy = tf.distribute.MirroredStrategy()

# 在分布式環境下訓練模型
with strategy.scope():
  train_dataset = strategy.experimental_distribute_dataset(train_dataset)
  test_dataset = strategy.experimental_distribute_dataset(test_dataset)
  model.compile(optimizer=optimizer, loss=loss_fn, metrics=["accuracy"])
  model.fit(train_dataset, epochs=10, validation_data=test_dataset)