the error cNMF

[zhoushican]

cnmf_obj.consensus(k=selected_K, 
                   density_threshold=density_threshold, 
                   show_clustering=True, 
                   close_clustergram_fig=False)

error

{
	"name": "TypeError",
	"message": "H should have the same dtype as X. Got H.dtype = float32.",
	"stack": "---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[26], line 1
----> 1 cnmf_obj.consensus(k=selected_K, 
      2                    density_threshold=density_threshold, 
      3                    show_clustering=True, 
      4                    close_clustergram_fig=False)

File ~/mambaforge/envs/omicverse_gpu/lib/python3.10/site-packages/omicverse/single/_cnmf.py:838, in cNMF.consensus(self, k, density_threshold, local_neighborhood_size, show_clustering, skip_density_and_return_after_stats, close_clustergram_fig, refit_usage)
    835     spectra_tpm_rf = spectra_tpm.loc[:,hvgs]
    837     spectra_tpm_rf = spectra_tpm_rf.div(tpm_stats.loc[hvgs, '__std'], axis=1)
--> 838     rf_usages = self.refit_usage(norm_tpm.X, spectra_tpm_rf)
    839     rf_usages = pd.DataFrame(rf_usages, index=norm_counts.obs.index, columns=spectra_tpm_rf.index)                                                                  
    841 save_df_to_npz(median_spectra, self.paths['consensus_spectra']%(k, density_threshold_repl))

File ~/mambaforge/envs/omicverse_gpu/lib/python3.10/site-packages/omicverse/single/_cnmf.py:667, in cNMF.refit_usage(self, X, spectra)
    664 else:
    665     refit_nmf_kwargs.update(dict(n_components = spectra.shape[0], H = spectra, update_H = False))
--> 667 _, rf_usages = self._nmf(X, nmf_kwargs=refit_nmf_kwargs)
    668 if (type(X) is pd.DataFrame) and (type(spectra) is pd.DataFrame):
    669     rf_usages = pd.DataFrame(rf_usages, index=X.index, columns=spectra.index)

File ~/mambaforge/envs/omicverse_gpu/lib/python3.10/site-packages/omicverse/single/_cnmf.py:546, in cNMF._nmf(self, X, nmf_kwargs)
    535 def _nmf(self, X, nmf_kwargs):
    536     \"\"\"
    537     Parameters
    538     ----------
   (...)
    544 
    545     \"\"\"
--> 546     (usages, spectra, niter) = non_negative_factorization(X, **nmf_kwargs)
    548     return(spectra, usages)

File ~/mambaforge/envs/omicverse_gpu/lib/python3.10/site-packages/sklearn/utils/_param_validation.py:214, in validate_params.<locals>.decorator.<locals>.wrapper(*args, **kwargs)
    208 try:
    209     with config_context(
    210         skip_parameter_validation=(
    211             prefer_skip_nested_validation or global_skip_validation
    212         )
    213     ):
--> 214         return func(*args, **kwargs)
    215 except InvalidParameterError as e:
    216     # When the function is just a wrapper around an estimator, we allow
    217     # the function to delegate validation to the estimator, but we replace
    218     # the name of the estimator by the name of the function in the error
    219     # message to avoid confusion.
    220     msg = re.sub(
    221         r\"parameter of \\w+ must be\",
    222         f\"parameter of {func.__qualname__} must be\",
    223         str(e),
    224     )

File ~/mambaforge/envs/omicverse_gpu/lib/python3.10/site-packages/sklearn/decomposition/_nmf.py:1122, in non_negative_factorization(X, W, H, n_components, init, update_H, solver, beta_loss, tol, max_iter, alpha_W, alpha_H, l1_ratio, random_state, verbose, shuffle)
   1119 X = check_array(X, accept_sparse=(\"csr\", \"csc\"), dtype=[np.float64, np.float32])
   1121 with config_context(assume_finite=True):
-> 1122     W, H, n_iter = est._fit_transform(X, W=W, H=H, update_H=update_H)
   1124 return W, H, n_iter

File ~/mambaforge/envs/omicverse_gpu/lib/python3.10/site-packages/sklearn/decomposition/_nmf.py:1670, in NMF._fit_transform(self, X, y, W, H, update_H)
   1663     raise ValueError(
   1664         \"When beta_loss <= 0 and X contains zeros, \"
   1665         \"the solver may diverge. Please add small values \"
   1666         \"to X, or use a positive beta_loss.\"
   1667     )
   1669 # initialize or check W and H
-> 1670 W, H = self._check_w_h(X, W, H, update_H)
   1672 # scale the regularization terms
   1673 l1_reg_W, l1_reg_H, l2_reg_W, l2_reg_H = self._compute_regularization(X)

File ~/mambaforge/envs/omicverse_gpu/lib/python3.10/site-packages/sklearn/decomposition/_nmf.py:1202, in _BaseNMF._check_w_h(self, X, W, H, update_H)
   1200 _check_init(H, (self._n_components, n_features), \"NMF (input H)\")
   1201 if H.dtype != X.dtype:
-> 1202     raise TypeError(
   1203         \"H should have the same dtype as X. Got H.dtype = {}.\".format(
   1204             H.dtype
   1205         )
   1206     )
   1207 # 'mu' solver should not be initialized by zeros
   1208 if self.solver == \"mu\":

TypeError: H should have the same dtype as X. Got H.dtype = float32."
}

https://omicverse.readthedocs.io/en/latest/Tutorials-single/t_cnmf/ https://github.com/dylkot/cNMF/issues/10