Vertex AI workbench and Google cloud storage problems accesing files

Google Cloud Build with AI Custom ML & MLOps
1

I am trying to create a ML project in which the job is a classification task of videos, so I upload those videos in Google cloud storage, and then I create a notebook on the workbench of vertex AI, for making data balancing, and then train my respective ML algorithm. But I have this problem:

  1. I want to use the video files from GCS without the need of downloading them again(that was the purpose of uploading them in GCS), but I don’t know how can i do this?.

I also try uploading de videos into the dataset space of the vertex AI workbench but still don’t know how to acces to this files without downloading them again.

2

If I’m not mistaken, you simply need to provide a schema file to set the structure and then provide the json input file with the right paths according to the schema file you’ve uploaded.
What do you mean by this?:

without the need of downloading them again

I’m not aware of any requirement to download a file in order to use it in Vertex AI, the learning algorithm should access the files on Cloud Storage and use them as needed.