Vertex AI training pipeline fails for unknown reasons

I tried to create a model using a dataset, image object detection, but it fails for an unknown reason.

Number constraints such as number of labels, number of images, etc.

Also, are there any restrictions on the characters that can be used in labels?

I can’t get detailed logs either.

Reducing the 3000 labels by half h to 1500 does not solve the problem.

Leave a Comment