Tokenizing .loom or .h5ad single cell RNA-seq data to rank value encoding .dataset format meeting quesrion

[DM0815]

when I perform following codes using my dataset, which is a very simple h5ad file, just contains "n_counts", "n_genes" in obs attribute, "ensembl_id","n_cells" in var attribute.
tk = TranscriptomeTokenizer(nproc=1)
tk.tokenize_data("./qin_data/stDiff-sc_filtered/",
"output/stDiff-sc/",
"./",
file_format="h5ad")

It reminds that
ValueError Traceback (most recent call last)

--> 507 raise ValueError("No objects to concatenate")
509 if keys is None:
510 objs_list = list(com.not_none(*objs_list))

ValueError: No objects to concatenate

Can you give me some suggestions? Thanks