Skip to content

TensorFlow-extended: How to generate interactive HTML outputs on databricks notebook #2221

@ghost

Description

I am trying to run some example codes at https://www.tensorflow.org/tfx/tutorials/model_analysis/tfma_basic on a databricks notebook.

I would like to use TensorFlow-extended to do ML model development and production on databricks GPU clusters.

The expected results from the code should be an interactive HTML file. There will be some tabs and buttons on the HTML. The users can manipulate the results and visualize them in multiple layouts.

I have run the code by Colab and it works well. You can try it by clicking the button "Run in google Colab" at the top of the webpage and run the notebook on Colab and see the interactive results.

But, when I run the same code on databricks notebook, no interactive results are generated. Only plain txt messages are dumped to the output cells in the notebook.

The databricks env :

7.1 ML Spark 3.0.0 Scala 2.12 GPU

Could you please help us about this ?

Metadata

Metadata

Assignees

Type

No type
No fields configured for issues without a type.

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions