Tensorboard 2.1.0 example on (https://www.tensorflow.org/tensorboard/get_started) not working

Question

I am using Win10Pro, gpu (CUDA 10.1), Python 3.7.5, Tensorflow 2.1.0 and Tensorboard 2.1.0 to run the tensorboard example given on (https://www.tensorflow.org/tensorboard/get_started) using the following code in ipython (after running %load_ext tensorboard in ipython console):

import tensorflow as tf
import datetime
import os
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

def create_model():
  return tf.keras.models.Sequential([
    tf.keras.layers.Flatten(input_shape=(28, 28)),
    tf.keras.layers.Dense(512, activation='relu'),
    tf.keras.layers.Dropout(0.2),
    tf.keras.layers.Dense(10, activation='softmax')
  ])
model = create_model()
model.compile(optimizer='adam',
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy'])

log_dir=os.path.join("logs","fit", datetime.datetime.now().strftime("%Y%m%d-%H%M%S"))
tensorboard_callback = tf.keras.callbacks.TensorBoard(log_dir=log_dir, histogram_freq=1)

model.fit(x=x_train, 
          y=y_train, 
          epochs=5, 
          validation_data=(x_test, y_test), 
          callbacks=[tensorboard_callback])

This runs successfully. Afterwards, I typed (%tensorboard --logdir logs/fit) in command prompt and I got the following message:

2020-01-24 11:40:14.612500: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll W0124 11:40:16.549416 8508 plugin_event_accumulator.py:294] Found more than one graph event per run, or there was a metagraph containing a graph_def, as well as one or more graph events. Overwriting the graph with the newest event. Serving TensorBoard on localhost; to expose to the network, use a proxy or pass --bind_all TensorBoard 2.1.0 at http://localhost:6006/ (Press CTRL+C to quit).

However, when I go the aforementioned address, there is nothing to see. My log directory has an absolute path: C:\Users\ak39.MEDMA\codes\logs\fit. Can anyone please help me in visualizing the Tensorboard?

Dieter Maes Dieter Maes · Accepted Answer · 2020-01-24T11:17:21

Can you try the following: Add the import at the top

from tensorflow.keras.callbacks import TensorBoard, LearningRateScheduler

add an update frequency to the callback

tensorboard_callback = TensorBoard(
  log_dir=log_dir,
  update_freq='batch',
  histogram_freq=1)

Also in the tensorflow guide (https://www.tensorflow.org/tensorboard/get_started) it is advised to start the tensorboard without the % when running it in the command line.

tensorboard --logdir logs/fit

If that does not work, then you could also try saving the checkpoints after each epoch by adding a checkpoint callback:

checkpoint_callback = ModelCheckpoint(
    filepath, monitor='val_accuracy', verbose=1,
    save_best_only=False, save_weights_only=False,
    save_frequency=1)

example:

model.fit(train_dataset,
          epochs=3, callbacks=[checkpoint_callback],
          validation_data=test_dataset,
          validation_freq=1)

Tensorboard 2.1.0 example on (https://www.tensorflow.org/tensorboard/get_started) not working

1 Answers