The predict function of GP

Questions
3

Thanks for your prompt reply.
The full error message is:

Traceback (most recent call last):

  File "C:\Users\xiaoyun\.spyder-pytorch\MCMC_IDDES.py", line 85, in <module>
    sys.exit(main())

  File "C:\Users\xiaoyun\.spyder-pytorch\MCMC_IDDES.py", line 75, in main
    mu,var = gp.predict(Z, point=point)

  File "C:\Anaconda3\lib\site-packages\pymc3\gp\gp.py", line 556, in predict
    mu, cov = self.predictt(Xnew, diag, pred_noise, given)

  File "C:\Anaconda3\lib\site-packages\pymc3\gp\gp.py", line 579, in predictt
    mu, cov = self._build_conditional(Xnew, pred_noise, diag, *givens)

  File "C:\Anaconda3\lib\site-packages\pymc3\gp\gp.py", line 478, in _build_conditional
    v = solve_lower(L, rxx)

  File "C:\Anaconda3\lib\site-packages\theano\graph\op.py", line 253, in __call__
    compute_test_value(node)

  File "C:\Anaconda3\lib\site-packages\theano\graph\op.py", line 130, in compute_test_value
    required = thunk()

  File "C:\Anaconda3\lib\site-packages\theano\graph\op.py", line 476, in rval
    r = p(n, [x[0] for x in i], o)

  File "C:\Anaconda3\lib\site-packages\theano\tensor\slinalg.py", line 251, in perform
    rval = scipy.linalg.solve_triangular(A, b, lower=True)

  File "C:\Anaconda3\lib\site-packages\scipy\linalg\basic.py", line 338, in solve_triangular
    raise ValueError('shapes of a {} and b {} are incompatible'

ValueError: shapes of a (88, 88) and b (400, 88) are incompatible

The version of pymc3 and the theano-pymc are 3.11.2 and 1.1.2.
Thanks again.