Datasets not found

[icyjayd]

I have installed this package and but I can't load the datasets.

My code is as follows:

from genomic_benchmarks.data_check import list_datasets
from genomic_benchmarks.dataset_getters.pytorch_datasets import get_dataset
from genomic_benchmarks.data_check import info
from genomic_benchmarks.loc2seq import download_dataset

When trying to download, for example, 'demo_coding_vs_intergenomic_seqs' I get FileNotFoundError: Dataset demo_coding_vs_intergenomic_seqs not found.

For completion's sake, I wrote code to attempt to download each of the datasets.

for dset in list_datasets():
    try:
        get_dataset(dset, split='train')
        print("success!")
    except:
        print(dset, "not found")

The output is as follows:

demo_coding_vs_intergenomic_seqs not found
human_enhancers_cohn not found
human_ocr_ensembl not found
demo_human_or_worm not found
human_ensembl_regulatory not found
drosophila_enhancers_stark not found
dummy_mouse_enhancers_ensembl not found
human_enhancers_ensembl not found
human_nontata_promoters not found

The same occurs with the info and download_dataset functions as well. Any help on what I'm doing wrong would be appreciated.

[ychuest]

May I ask how you solved this problem? I'm having the same problem.

[katarinagresova]

Dear @icyjayd and @ychuest, thank you for using Genomic Benchmarks,

this looks like the package is trying to find information about the dataset, but failed. When dataset is not downloaded yet, the package requests following url for metadata - https://raw.githubusercontent.com/ML-Bioinfo-CEITEC/genomic_benchmarks/main/datasets/human_nontata_promoters/metadata.yaml. Can you make sure you are able to access this url from the computer you are running the code?