AttributeError: module 'tensorflow' has no attribute 'Summary'

[palunel]

I use pytorch with the following implementation:

from tensorboardcolab import TensorBoardColab
tb = TensorBoardColab()

and this call:

tb.save_value('Validation Accuracy', 'valid_acc',i, validation_acc)

Traceback gives me this error:

AttributeError                            Traceback (most recent call last)

<ipython-input-23-1106bec90267> in <module>()
      6 optimizer = optim.Adam(model.parameters(), lr=0.001, weight_decay=0.0001)
      7 training_epochs = 150
----> 8 metrics = train(model, train_loader, test_loader, loss, optimizer, training_epochs)
      9 
     10 summary(model, (3, 32,32))

1 frames

/usr/local/lib/python3.6/dist-packages/tensorboardcolab/core.py in save_value(self, graph_name, line_name, epoch, value)
     99                 tf.contrib.summary.scalar(graph_name, value)
    100         else:
--> 101             summary = tf.Summary()
    102             summary_value = summary.value.add()
    103             summary_value.simple_value = value

AttributeError: module 'tensorflow' has no attribute 'Summary'

[denabazazian]

In the recent version of TensorFlow, you should replace these lines:

summary = tf.Summary()
summary_value = summary.value.add()
summary_value.simple_value = value
summary_value.tag = graph_name

with:

with self.writer.as_default():

              tf.summary.scalar(graph_name, value)

              self.writer.flush()

Also, replace this line:

self.writer = tf.summary.FileWriter(log_dir)

with:

self.writer = tf.summary.create_file_writer(log_dir)

[Michael-J98]

I got another error after following the above instruction：

ValueError Traceback (most recent call last)
in ()
19 if x%10==0:
20 print("Epoch[{}], Train_Iterations[{}]--mean_loss:{} ".format(epoch,x,loss))
---> 21 tb.save_value('Train Loss', 'train_loss', global_iter, loss.item())
22 optimizer.zero_grad()
23 loss.backward()

2 frames
/usr/local/lib/python3.6/dist-packages/tensorboardcolab/core.py in save_value(self, graph_name, line_name, epoch, value)
100 else:
101 with self.writer.as_default():
--> 102 tf.summary.scalar(graph_name, value)
103 self.writer.flush()
104

/usr/local/lib/python3.6/dist-packages/tensorboard/plugins/scalar/summary_v2.py in scalar(name, data, step, description)
66 tensor=tf.cast(data, tf.float32),
67 step=step,
---> 68 metadata=summary_metadata,
69 )
70

/usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/summary_ops_v2.py in write(tag, tensor, step, metadata, name)
647 step = get_step()
648 if step is None:
--> 649 raise ValueError("No step set via 'step' argument or "
650 "tf.summary.experimental.set_step()")
651 if metadata is None:

ValueError: No step set via 'step' argument or tf.summary.experimental.set_step()

Is this about the version too?

[YaraAlnaggar]

I got another error after following the above instruction：

ValueError Traceback (most recent call last)
in ()
19 if x%10==0:
20 print("Epoch[{}], Train_Iterations[{}]--mean_loss:{} ".format(epoch,x,loss))
---> 21 tb.save_value('Train Loss', 'train_loss', global_iter, loss.item())
22 optimizer.zero_grad()
23 loss.backward()

2 frames
/usr/local/lib/python3.6/dist-packages/tensorboardcolab/core.py in save_value(self, graph_name, line_name, epoch, value)
100 else:
101 with self.writer.as_default():
--> 102 tf.summary.scalar(graph_name, value)
103 self.writer.flush()
104

/usr/local/lib/python3.6/dist-packages/tensorboard/plugins/scalar/summary_v2.py in scalar(name, data, step, description)
66 tensor=tf.cast(data, tf.float32),
67 step=step,
---> 68 metadata=summary_metadata,
69 )
70

/usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/summary_ops_v2.py in write(tag, tensor, step, metadata, name)
647 step = get_step()
648 if step is None:
--> 649 raise ValueError("No step set via 'step' argument or "
650 "tf.summary.experimental.set_step()")
651 if metadata is None:

ValueError: No step set via 'step' argument or tf.summary.experimental.set_step()

Is this about the version too?

Have met the same error and adding step argument which indicates batch/epoch number to tf.summary.scalar(graph_name, value) solves the issue.
tf.summary.scalar(graph_name, value) to tf.summary.scalar(graph_name, value,step=step)

[pvhdhxd]

In the recent version of TensorFlow, you should replace these lines:

summary = tf.Summary()
summary_value = summary.value.add()
summary_value.simple_value = value
summary_value.tag = graph_name

with:

with self.writer.as_default():

tf.summary.scalar(graph_name, value)

self.writer.flush()

Also, replace this line:

self.writer = tf.summary.FileWriter(log_dir)

with:

self.writer = tf.summary.create_file_writer(log_dir)

Dear denabazazian,
I followed to your guidance, however I met an error:

~\AppData\Roaming\Python\Python38\site-packages\keras\callbacks.py in on_epoch_end(self, epoch, logs)
902 #.
903 # Doi self.write to writer
--> 904 with self.writer.as_default():
905 tf.summary.scalar(graph_name, value)
906 self.writer.flush()

AttributeError: 'FileWriter' object has no attribute 'as_default'

I am using python 3.8.5 and tensorflow 2.5.0

Could anyone help me to fix it