DenseHMM in pomegranate 1.0.0 doesn't work with single-element sequences

[vidraj]

In pomegranate 1.0.0, an attempt to classify a single-element sequence using DenseHMM fails with a RuntimeError due to an illegal reshaping operation. SparseHMM works fine.

Sample code that reproduces the problem (mostly copied from the docs):

import numpy as np
from pomegranate.distributions import Categorical
from pomegranate.hmm import DenseHMM

d1 = Categorical([[0.25, 0.25, 0.25, 0.25]])
d2 = Categorical([[0.10, 0.40, 0.40, 0.10]])

model = DenseHMM()
model.add_distributions([d1, d2])
model.add_edge(model.start, d1, 0.5)
model.add_edge(model.start, d2, 0.5)
model.add_edge(d1, d1, 0.9)
model.add_edge(d1, d2, 0.1)
model.add_edge(d2, d1, 0.1)
model.add_edge(d2, d2, 0.9)

sequence = 'A'
X = np.array([[[['A', 'C', 'G', 'T'].index(char)] for char in sequence]])

y_hat = model.predict(X)
print(y_hat)

Expected output:

tensor([[0]])

Actual output:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "venv/lib/python3.11/site-packages/pomegranate/hmm/_base.py", line 518, in predict
    return torch.argmax(self.predict_log_proba(X, priors=priors), dim=-1)
                        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "venv/lib/python3.11/site-packages/pomegranate/hmm/_base.py", line 459, in predict_log_proba
    _, r, _, _, _ = self.forward_backward(X, priors=priors)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "venv/lib/python3.11/site-packages/pomegranate/hmm/dense_hmm.py", line 486, in forward_backward
    t = t.reshape(n, l-1, -1)
        ^^^^^^^^^^^^^^^^^^^^^
RuntimeError: cannot reshape tensor of 0 elements into shape [1, 0, -1] because the unspecified dimension size -1 can be any value and is ambiguous