Chansung Park

So we could convert prediction results into pandas DataFrame, an run TFMA on that I guess?

sorry for the long late response. you can just call the function like below `cifar10_dataset = Cifar10()`

@kkweon I love this Is this something to be set up during a GitHub Action? or manually whenever before pushing?

AFAIK, FSDP+QDoRA is not supported feature in HF official releases like transformers, peft, ...

Thank you @shabie I think it could be common to have a test dataset in a single repo while we could have training dataset from multiple sources. At least this...

@JIElite AFAIK, the ratio doesn't have any impact on the test split.

@sayakpaul I just invited her as a collaborator to this repository!

About this example of using TensorBoard(https://huggingface.co/keras-io/lowlight-enhance-mirnet/tensorboard), can we automatically generate Model card based on the TensorBoard logs?

also, it is worth noting how TFX official `ModelCardGenerator` works and what it generates. - https://github.com/digits/tfx-addons/blob/add-model-card-component/examples/model_card_generator/MLMD_Model_Card_Toolkit_Demo.ipynb

There are some things to note about TensorBoard. - The logs of TensorBoard does not represent the final performance. It comes with many logs including lots of experiments. - I...