深度学习（33）随机梯度下降十一: TensorBoard可视化

TensorBoard

• Installation
• Curves
• Image Visualization
  Installation

pip install tensorboard

Principle

• Listen logdir
  监听目录
• build summary instance
  新建一个日志
• fed data into summary instance
  将数据送入日志

Step1. run listener

进入需要监听的文件夹

cd /Users/xuruihang/Documents/深度学习

启动

tensorboard –logdir logs

如图所示:

进入http://localhost:6006/，如图所示:

Step2. build summary

current_time = datetime.datetime.now().strftime(“%Y%m%d-%H%M%S”)
log_dir = ‘logs/’ + current_time
summary_writer = tf.summary.create_file_writer(log_dir)

其中，log_dir为监听文件的路径。

Step3.1 fed scalar（监听标量）

with summary_writer.as_default():
	tf.summary.scalar(‘loss’, float(loss), step=epoch)
tf.summary.scalar(‘accuracy’, float(train_accuracy), step=epoch)

其中，step默认为x轴。

Step3.2 fed single Image（监听单张图片）

# get x from (x,y)
sample_img = next(iter(db))[0]
# get first image instance
sample_img = sample_img[0]
sample_img = tf.reshape(sample_img, [1, 28, 28, 1])
with summary_writer.as_default():
	tf.summary.image(“Training sample:”, sample_img, step=0)

如图所示:

Step3.3 fed multi-images（监听多张图片）

val_images = x[:25]
val_images = tf.reshape(val_images, [-1, 28, 28, 1])
with summary_writer.as_default():
	tf.summary.scalar(‘test-acc’, float(loss), step=step)
	tf.summary.image(“val-onebyone-images:”, val_images, max_outputs=25, step=step)

如图所示:

Step3.4 将多张图片组合为一张图片:

val_images = tf.reshape(val_images, [-1, 28, 28])
figure = image_grid(val_images)
tf.summary.image(‘val-images:’, plot_to_image(figure), step=step)

如图所示:

参考文献:
[1] 龙良曲:《深度学习与TensorFlow2入门实战》