Step #4: TensorBoard

Synthetic Data Generator for Tabular Data (Latest)

Now that you have successfully pretrained your model, the next step is to visualize and monitor the training using metrics such as loss/train per epoch. As you may recall, model checkpoint and tensorboard events were saved. In the next step, we will go over training curves for dataset validation.

Within your Jupyter lab tab, open the Tensorboard notebook and run it.

Once you are done running through the notebook, proceed with Step #5.

© Copyright 2022-2023, NVIDIA. Last updated on Jan 10, 2023.