Dataset10X¶
-
class
scvi.dataset.Dataset10X(dataset_name=None, filename=None, save_path='data/10X', url=None, type='filtered', dense=False, measurement_names_column=1, remove_extracted_data=False, delayed_populating=False)[source]¶ Bases:
scvi.dataset.dataset.DownloadableDatasetLoads a file from 10x website.
- Parameters
dataset_name (
str,NoneOptional[str]) – Name of the dataset file. Has to be one of: “frozen_pbmc_donor_a”, “frozen_pbmc_donor_b”, “frozen_pbmc_donor_c”, “fresh_68k_pbmc_donor_a”, “cd14_monocytes”, “b_cells”, “cd34”, “cd56_nk”, “cd4_t_helper”, “regulatory_t”, “naive_t”, “memory_t”, “cytotoxic_t”, “naive_cytotoxic”, “pbmc8k”, “pbmc4k”, “t_3k”, “t_4k”, “neuron_9k”, “pbmc_1k_protein_v3”, “pbmc_10k_protein_v3”, “malt_10k_protein_v3”, “pbmc_1k_v2”, “pbmc_1k_v3”, “pbmc_10k_v3”, “hgmm_1k_v2”, “hgmm_1k_v3”, “hgmm_5k_v3”, “hgmm_10k_v3”, “neuron_1k_v2”, “neuron_1k_v3”, “neuron_10k_v3”, “heart_1k_v2”, “heart_1k_v3”, “heart_10k_v3”.filename (
str,NoneOptional[str]) – manual override of the filename to write to.save_path (
strstr) – Location to use when saving/loading the data.url (
str,NoneOptional[str]) – manual override of the download remote location. Note that we already provide urls for most 10X datasets, which are automatically formed only using thedataset_name.dense (
boolbool) – Whether to load as dense or sparse. If False, data is cast to sparse usingscipy.sparse.csr_matrix.measurement_names_column (
intint) – column in which to find measurement names in the corresponding .tsv file.remove_extracted_data (
boolbool) – Whether to remove extracted archives after populating the dataset.
Examples
>>> tenX_dataset = Dataset10X("neuron_9k")
Methods Summary
Returns exact path for the data in the archive.
populate()Populates a
DonwloadableDatasetobject’s data attributes.Methods Documentation