Tensorflow_MNIST

MNIST dataset

1.Summarization

2.loading

import tensorflow as tf
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

model = tf.keras.models.Sequential([
  tf.keras.layers.Flatten(),
  tf.keras.layers.Dense(512, activation=tf.nn.relu),
  tf.keras.layers.Dropout(0.2),
  tf.keras.layers.Dense(10, activation=tf.nn.softmax)
])
model.compile(optimizer='adam',
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy'])

model.fit(x_train, y_train, epochs=5)
model.evaluate(x_test, y_test)

Run_IN_A_CO_NOTEBOOK

the Result

Cloud TPU

Tensor Processing Unit

It is a ASIC specially designed for machine learning and TensorFlow customization (integrated circuit chip technology). The TPU is a programmable AI accelerator that provides high throughput, low precision calculations (such as 8 bits), oriented to use or run models rather than training models.

它是一个专门为机器学习和TensorFlow定制的ASIC（集成电路芯片技术）。TPU是一个可编程的人工智能加速器，提供高吞吐量的低精度计算（如8位），面向使用或运行模型而不是训练模型。

The Unknown Word

The First Column	The second Column
domain-specific handware	领域定制硬件
TPU	Tensor Processing Unit张量处理单元
Tensor	张量，代表了N维数组
Flow	流，代表了基于数据流图的计算
customization	定制
low precision	低精度
precision	[pri'sigen]精度

转载于:https://www.cnblogs.com/hugeng007/p/9498541.html