[BUG] image in evaluation table

[nathalie321]

Willingness to contribute

Yes. I would be willing to contribute a fix for this bug with guidance from the MLflow community.

MLflow version

2.12.1

System information

• OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Ubuntu 22.04
• Python version: 3.11
• yarn version, if running the dev UI:

Describe the problem

log_table support for image does not display the image correctly.

Steps to reproduce the bug

Sample from docs doesn't display image.

image = mlflow.Image([[1, 2, 3]])
table_dict = {
"inputs": ["Show me a dog", "Show me a cat"],
"outputs": [image, image],
}
from mlflow import MlflowClient
df = pd.DataFrame.from_dict(table_dict)
with mlflow.start_run() as run:
client = MlflowClient()
client.log_table(run.info.run_id, data=df, artifact_file="image_gen.json")

Code to generate data required to reproduce the bug

No response

Is the console panel in DevTools showing errors relevant to the bug?

No response

Does the network panel in DevTools contain failed requests relevant to the bug?

No response

[WeichenXu123]

Hello,

The image is visible well in run details > artifacts,
but I am not sure this is bug or by design in evaluation table.

I will ask our team members and get back to you :)

[nathalie321]

Ah yes, I can clearly see it inside the artifacts.
I'm also interested in it being displayed in evaluation table, thank you!

[WeichenXu123]

@nathalie321 Yes this is a bug. Feel free to file a PR to fix it. Appreciate your reporting and contribution!

[WeichenXu123]

thanks for reporting this, I talked with our team folks, this is a bigger project than just a backend change - REACT modules are needed, we plan to fix during Q2. please stay tuned. :)

[github-actions]

@mlflow/mlflow-team Please assign a maintainer and start triaging this issue.

[WeichenXu123]

After code sync from our internal repository #12026 , this issue should be addressed.