Unable to see plots, metrics etc of Evaluator component

I am unable to see any outputs from Evaluator component.

Code used →
context.show( evaluator.outputs[‘evaluation’ ])

Getting output as

Artifact at tfx/Evaluator/Evaluation/94

Error displaying widget: model not found

Are you in a notebook using the InteractiveContext ? What does the artifact explorer show?

Yes, I am using Interactive context

Artifact explorer shows the following

Outputs shows the following

Sorry for the image quality, can’t take a screenshot due to company policy

In the folder structure
I am able to see various tf record files such as, attributes, metrics, plots, validations .

But am unable to visualise these files.

@Yasir_Modak for visibility


You can visualize the EvalResult from evaluator component by loading it using TFMA library and plot them using tfma.view.render_slicing_metrics as shown in documentation. Let us know if you face any issues while implementation. Thank you!

Example Code:

import tensorflow_model_analysis as tfma

output_path = evaluator.outputs['evaluation'].get()[0].uri

# Load the evaluation results.
eval_result = tfma.load_eval_result(output_path)

# Visualize the metrics and plots using tfma.view.render_slicing_metrics,
# tfma.view.render_plot, etc.

# Load the validation results
validation_result = tfma.load_validation_result(output_path)


What’s tfma_result ?

tfma_result is not defined