CYorick

Hi, is the problem solved? I also met the problem

> Hi @CYorick > > It might be memory related. The code in the development branch is more memory friendly. > > See #202 on how to use it. >...

The problem can be solved by setting the "ray_n_cpu" as None

That's weird, I can run get_search_space and merge_cistromes, but got stuck in calculate_regions_to_genes_relationships. Thanks for your help. Best, Yorick

Here is the description of my SCENIC+ object SCENIC+ object with n_cells x n_genes = 940 x 23353 and n_cells x n_regions = 940 x 66831 metadata_regions:'Chromosome', 'Start', 'End', 'Width',...

I finally find out that the EXP_df = scplus_obj.to_df(layer='EXP') with this gene has two columns, I am not sure why it happens, maybe we should simply choose the first column?

> Hi @sid5427 > > From the error I suspect that `region_sets['DARs']` might be empty or contain empty entries. Could you show the output of `region_sets['DARs']` to confirm this? >...

This is my code: ``` import torch from transformers import AutoTokenizer, AutoModelForCausalLM # Load the pretrained HyenaDNA model and tokenizer checkpoint = 'LongSafari/hyenadna-large-1m-seqlen-hf' cache_dir = '/blue/sai.zhang/chen.yongzhuo/LLM_evaluation/cache' tokenizer = AutoTokenizer.from_pretrained(checkpoint, trust_remote_code=True,...