慢tensorflow代码，是否可以批量评估并获取多个损失分数？

import tensorflow as tf

# 定义模型
model = tf.keras.Sequential([
    tf.keras.layers.Dense(64, activation='relu'),
    tf.keras.layers.Dense(10, activation='softmax')
])

# 定义损失函数
loss_fn = tf.keras.losses.SparseCategoricalCrossentropy()

# 定义优化器
optimizer = tf.keras.optimizers.Adam()

# 定义评估指标
metrics = ['accuracy']

# 编译模型
model.compile(optimizer=optimizer, loss=loss_fn, metrics=metrics)

# 加载数据集
dataset = tf.data.Dataset.from_tensor_slices((x_train, y_train)).batch(batch_size)

# 批量评估并获取损失分数
loss_scores = []
for batch_data in dataset:
    x_batch, y_batch = batch_data
    loss, _ = model.evaluate(x_batch, y_batch, verbose=0)
    loss_scores.append(loss)

print(loss_scores)