UnboundLocalError: local variable 'dist' referenced before assignment

Question

I am trying to train a model for supervised learning for Hidden Markov Model (HMM)and test it on a set of observations however, keep getting this error. The goal is to predict the state based on the observations. How can I fix this and how can I view the transition matrix?

The version for Pomegranate is 0.14.4 Trying this from the source: https://github.com/jmschrei/pomegranate/issues/1005

from pomegranate import *
import numpy as np
# Supervised method that calculates the transition matrix:
d1 = State(UniformDistribution.from_samples([3.243221498397177, 3.210684537495482, 3.227662201472816,
    3.286410817416738, 3.290573650708864, 3.286058136226862, 3.266480693857006]))
d2 = State(UniformDistribution.from_samples([3.449282367485096, 1.97317859465635, 1.897551432353011,
     3.454609351559659, 3.127357456033111, 1.779308337786426, 3.802891929694426, 3.359766157565077, 2.959428499979418]))
d3 = State(UniformDistribution.from_samples([1.892812118441474, 1.589353118681066, 2.09269978285637,
     2.104391496570218, 1.656771181054144]))
model = HiddenMarkovModel()
model.add_states(d1, d2, d3)
# print(model.to_json())
model.bake()
model.fit([3.2, 6.7, 10.55], labels=[1, 2, 3], algorithm='labeled')
all_pred = model.predict([2.33, 1.22, 1.4, 10.6])

Error:

 File "C:\Program Files\JetBrains\PyCharm Community Edition 2021.2\plugins\python-ce\helpers\pydev\_pydev_bundle\pydev_umd.py", line 198, in runfile
    pydev_imports.execfile(filename, global_vars, local_vars)  # execute the script
  File "C:\Program Files\JetBrains\PyCharm Community Edition 2021.2\plugins\python-ce\helpers\pydev\_pydev_imps\_pydev_execfile.py", line 18, in execfile
    exec(compile(contents+"\n", file, 'exec'), glob, loc)
  File "C:/Users/", line 774, in <module>
    model.bake()
  File "pomegranate/hmm.pyx", line 1047, in pomegranate.hmm.HiddenMarkovModel.bake
UnboundLocalError: local variable 'dist' referenced before assignment

That's a bug in the `pomegranate` library. Talk to its authors, not to us. — Charles Duffy, Nov 28 '22 at 16:48
...that said, if I _were_ to run this down inside `pomegranate`, I'd want to know exactly which version of the library you're running so I'm not wasting my time looking at the wrong source. It can't be the latest one -- [in the tip of the main branch, line 1047 is a comment](https://github.com/jmschrei/pomegranate/blob/f115a242a5b50854bbf199d43fe2cfd061e9708a/pomegranate/hmm.pyx#L1047). — Charles Duffy, Nov 28 '22 at 16:49
(...that said, from a cursory glance at the source, and making some assumptions about which version you're running, it _looks_ like you could get this bug by having only silent states -- that is to say, states with `distribution=None`. Still, it's a pomegranate problem that they don't raise a useful/meaningful exception in that case) — Charles Duffy, Nov 28 '22 at 16:54
Duplicate of this issue on the source repo: https://github.com/jmschrei/pomegranate/issues/1005 Actually, the code there is identical to the code here. @maximus, are you smalik89 on GitHub? If so, it would be helpful to spell that out explicitly in the question, and if not, please quote your sources. — joanis, Nov 28 '22 at 17:29

score 1 · Answer 1 · answered Dec 07 '22 at 11:37

Seems like a bug in promegranate.

This Python UnboundLocalError: Local variable referenced before assignment" occurs when we reference a local variable before assigning a value to it in a function.

Observing the hmm.pyx code, dist is assigned a value in a function's body(which is a local variable) and not declared as global. To solve the error, mark the variable as global in the function definition,

e.g. global dist

Try adding dist variable in your code that you provided. For e.g. Hope this helps!

UnboundLocalError: local variable 'dist' referenced before assignment

1 Answers1