How to use the scrapbook.glue function in scrapbook

# In[19]:


# The following code is used by the notebook "24_run_notebook_on_azureml.ipynb" to log metrics when using papermill or scrapbook
# to run this notebook. We can comment out this cell when we are running this notebook directly.

training_losses = [x.numpy().ravel()[0] for x in learn.recorder.losses]
training_accuracy = [x[0].numpy().ravel()[0] for x in learn.recorder.metrics]

# pm.record may get deprecated and completely replaced by sb.glue:
# https://github.com/nteract/scrapbook#papermills-deprecated-record-feature
try:
    sb.glue("training_loss", training_losses)
    sb.glue("training_accuracy", training_accuracy)
    sb.glue("Accuracy on validation set:", 100 * float(metric))
except Exception:
    pm.record("training_loss", training_losses)
    pm.record("training_accuracy", training_accuracy)
    pm.record("Accuracy on validation set:", 100 * float(metric))

check.inst_param(
            dagster_event, 'dagster_event', (Materialization, ExpectationResult, TypeCheck, Failure)
        )

        if not self.in_pipeline:
            return dagster_event

        # deferred import for perf
        import scrapbook

        event_id = 'event-{event_uuid}'.format(event_uuid=str(uuid.uuid4()))
        out_file_path = os.path.join(self.marshal_dir, event_id)
        with open(out_file_path, 'wb') as fd:
            fd.write(pickle.dumps(dagster_event, PICKLE_PROTOCOL))

        scrapbook.glue(event_id, out_file_path)

interp.plot_top_losses(9, figsize=(15, 11))


# In[19]:


# The following code is used by the notebook "24_run_notebook_on_azureml.ipynb" to log metrics when using papermill or scrapbook
# to run this notebook. We can comment out this cell when we are running this notebook directly.

training_losses = [x.numpy().ravel()[0] for x in learn.recorder.losses]
training_accuracy = [x[0].numpy().ravel()[0] for x in learn.recorder.metrics]

# pm.record may get deprecated and completely replaced by sb.glue:
# https://github.com/nteract/scrapbook#papermills-deprecated-record-feature
try:
    sb.glue("training_loss", training_losses)
    sb.glue("training_accuracy", training_accuracy)
    sb.glue("Accuracy on validation set:", 100 * float(metric))
except Exception:
    pm.record("training_loss", training_losses)
    pm.record("training_accuracy", training_accuracy)
    pm.record("Accuracy on validation set:", 100 * float(metric))

return value

        # deferred import for perf
        import scrapbook

        if not self.solid_def.has_output(output_name):
            raise DagstermillError(
                'Solid {solid_name} does not have output named {output_name}'.format(
                    solid_name=self.solid_def.name, output_name=output_name
                )
            )

        runtime_type = self.solid_def.output_def_named(output_name).runtime_type

        out_file = os.path.join(self.marshal_dir, 'output-{}'.format(output_name))
        scrapbook.glue(output_name, write_value(runtime_type, value, out_file))

# In[19]:


# The following code is used by the notebook "24_run_notebook_on_azureml.ipynb" to log metrics when using papermill or scrapbook
# to run this notebook. We can comment out this cell when we are running this notebook directly.

training_losses = [x.numpy().ravel()[0] for x in learn.recorder.losses]
training_accuracy = [x[0].numpy().ravel()[0] for x in learn.recorder.metrics]

# pm.record may get deprecated and completely replaced by sb.glue:
# https://github.com/nteract/scrapbook#papermills-deprecated-record-feature
try:
    sb.glue("training_loss", training_losses)
    sb.glue("training_accuracy", training_accuracy)
    sb.glue("Accuracy on validation set:", 100 * float(metric))
except Exception:
    pm.record("training_loss", training_losses)
    pm.record("training_accuracy", training_accuracy)
    pm.record("Accuracy on validation set:", 100 * float(metric))