Normalization issue on Stereo-seq tutorial

[tvegawaichman]

Hello! When I use an object that already has normalization in adata.X (following the Stereo-seq tutorial), the model runs the preprocessing:

def preprocess(adata):
    sc.pp.highly_variable_genes(adata, flavor="seurat_v3", n_top_genes=3000)
    sc.pp.normalize_total(adata, target_sum=1e4)
    sc.pp.log1p(adata)
    sc.pp.scale(adata, zero_center=False, max_value=10)

to get the highly variable genes. That's why I got the warning:
UserWarning: flavor='seurat_v3' expects raw count data, but non-integers were found. warnings.warn()

This means the new adata.X will be a "normalization" of the previously normalized adata.X, converting the count matrix twice.
Won't this affect the model and the results?