Uploaded Dataset not visible when using load_dataset

#4
by Simon1997 - opened
GEM benchmark org

Hi,

I am new to GEM and I am attempting to upload a new dataset that I have been using for my research. I followed the tutorials provided on the website (https://gem-benchmark.com/tutorials) and made sure that the dataset is in the correct format (CSV), the dataset splits are named correctly, and that the datacard is completed and added (I created the datacard with the tool provided).

After uploading, I can see the dataset is visible in the GEM Benchmark Space, as shown below

afbeelding.png

However, when I try to load the dataset and test it, I get the error that the dataset is not part of the GEM list.

Could someone help me understand why this is going wrong? Am I missing a step or have I made a mistake?

Kind regards,

Simon

Simon1997 changed discussion status to closed
GEM benchmark org

I can load it by passing the complete dataset name: "GEM/Elongated_CACAPO_for_E2E"

ds = load_dataset("GEM/Elongated_CACAPO_for_E2E")

Please note this discussion is taking place in another dataset called "gem".

GEM benchmark org

Thanks for checking Albert, I fixed an issue and could use the dataset, after which I closed the discussion. Apologies for placing the discussion in the wrong place.

Sign up or log in to comment