model.fit_transform valueerror expected 2D array, got scalar array instead

Question

I am trying to decompose a spectrum of a mixture of compounds into its main components ("pure compounds") by means of non-negative matrix factorization. mixed_spectrum is a np array expressing the spectrum of my compounds mixture, and W_init contains column vectors with spectra of known pure compounds and an extra component for unknown substances. My script for the analysis is the following:

import numpy as np
from sklearn.decomposition import NMF

pure_compounds /= np.max(pure_compounds)
n_components = len(pure_compounds) + 1
model = NMF(n_components=n_components, init='custom')
W_init = np.concatenate((pure_compounds, np.ones((pure_compounds.shape[0], 1))), axis=1)
print(W_init)
print(np.shape(W_init))
mix = mixed_spectrum.reshape(-1, 1)
print(mix)
print(np.shape(mix))
W = model.fit_transform(mix, W=W_init)
H = model.components_
for i in range(n_components):
    coeff = W[:, i] / np.sum(W, axis=1)
    print('Concentration coefficients for compound', i, ':', coeff)

which gives me the output

[[0.00000000e+00 2.76922293e-04 1.44273203e-03 0.00000000e+00
  1.00000000e+00]
 [0.00000000e+00 2.85985939e-04 1.40933148e-03 0.00000000e+00
  1.00000000e+00]
 [0.00000000e+00 2.54579755e-04 1.52763958e-03 0.00000000e+00
  1.00000000e+00]
 ...
 [1.00000000e-06 1.00000000e-06 1.00000000e-06 1.00000000e-06
  1.00000000e+00]
 [1.00000000e-06 1.00000000e-06 1.00000000e-06 1.00000000e-06
  1.00000000e+00]
 [1.00000000e-06 1.00000000e-06 1.00000000e-06 1.00000000e-06
  1.00000000e+00]]
(8954, 5)
[[2.31330983e-02]
 [1.00000000e-06]
 [1.61938395e-02]
 ...
 [2.42525353e-02]
 [1.95440454e-02]
 [2.98149646e-03]]
(8954, 1)

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
~\AppData\Local\Temp\ipykernel_6528\919808671.py in <module>
      8 print(mix)
      9 print(np.shape(mix))
---> 10 W = model.fit_transform(mix, W=W_init)
     11 H = model.components_
     12 for i in range(n_components):

~\Anaconda3\lib\site-packages\sklearn\decomposition\_nmf.py in fit_transform(self, X, y, W, H)
   1536 
   1537         with config_context(assume_finite=True):
-> 1538             W, H, n_iter = self._fit_transform(X, W=W, H=H)
   1539 
   1540         self.reconstruction_err_ = _beta_divergence(

~\Anaconda3\lib\site-packages\sklearn\decomposition\_nmf.py in _fit_transform(self, X, y, W, H, update_H)
   1595 
   1596         # initialize or check W and H
-> 1597         W, H = self._check_w_h(X, W, H, update_H)
   1598 
   1599         # scale the regularization terms

~\Anaconda3\lib\site-packages\sklearn\decomposition\_nmf.py in _check_w_h(self, X, W, H, update_H)
   1460         n_samples, n_features = X.shape
   1461         if self.init == "custom" and update_H:
-> 1462             _check_init(H, (self._n_components, n_features), "NMF (input H)")
   1463             _check_init(W, (n_samples, self._n_components), "NMF (input W)")
   1464             if H.dtype != X.dtype or W.dtype != X.dtype:

~\Anaconda3\lib\site-packages\sklearn\decomposition\_nmf.py in _check_init(A, shape, whom)
     52 
     53 def _check_init(A, shape, whom):
---> 54     A = check_array(A)
     55     if np.shape(A) != shape:
     56         raise ValueError(

~\Anaconda3\lib\site-packages\sklearn\utils\validation.py in check_array(array, accept_sparse, accept_large_sparse, dtype, order, copy, force_all_finite, ensure_2d, allow_nd, ensure_min_samples, ensure_min_features, estimator)
    759             # If input is scalar raise error
    760             if array.ndim == 0:
--> 761                 raise ValueError(
    762                     "Expected 2D array, got scalar array instead:\narray={}.\n"
    763                     "Reshape your data either using array.reshape(-1, 1) if "

ValueError: Expected 2D array, got scalar array instead:
array=None.
Reshape your data either using array.reshape(-1, 1) if your data has a single feature or array.reshape(1, -1) if it contains a single sample.

How can I fix the issue? I tried to play with the shape of the mix and W_init variables, but without success.

Answer 1

I solved the issue. When choosing a custom initialization for an NMF decomposition, one has to define both the initial W and H matrices.

Furthermore, one has to make sure that all arrays X, W and H are C-contiguous. If a numpy array has to be transposed, it is necessary to generate a .copy() to ensure this contiguousness.

My corrected code:

n_comps = len(pure_compounds[0]) + 1
model = NMF(n_components=n_comps, init='custom')
W_init = np.concatenate((pure_compounds, np.ones((pure_compounds.shape[0], 1))), axis=1).copy()
H_i = np.array(np.random.uniform(0.1, 1, size=n_comps))
H_init = H_i[:, None].copy()
mi = np.array(mixed_spectrum.copy())
mix = mi[:, None].copy()
W = model.fit_transform(mix, W=W_init, H=H_init)
H = model.components_
for i in range(n_components):
    coeff = H[i][0] / np.sum(H)
    print('Concentration coefficients for compound', i, ':', coeff)