Need "no code" help with buckets and datasets.

[croppers]

I have uploaded thousands of images (individual jpgs) to a Bucket. I now want to access those and use them in a new dataset. I can see all my uploaded images in the "cloud storage" "bucket" interface, but I cannot, for the life of me, work out how to attach any of those to a new dataset, except one at a time. Is there a way to make these images available to a new dataset in batches? Also, and I can't stress this enough, if that involves any form of coding, then that's a show-stopper for me.

[Joevanie]

To upload images to a dataset in batch, you may create a csv file containing all the images that you will upload. Here is the full documentation. In summary, below is what your csv file should contain:

[ML_USE],GCS_FILE_PATH,[LABEL]

Note that GCS_FILE_PATH is the only one required. If you only need to upload from bucket to dataset, you can just list all the filepaths like this:

gs://BUCKET_NAME/FILE_NAME_1.jpg
gs://BUCKET_NAME/FILE_NAME_2.jpg
gs://BUCKET_NAME/FILE_NAME_3.jpg

You can program this to save time but you can also do it manually if that's what you prefer.

After creating the csv file:

1. Upload it in the same bucket containing the images

2. Choose "select an import file from Cloud Storage" in Datasets

3. Choose the import csv file

If you need further clarification, here is the full documentation for Creating datasets. Hope this helps