I constructed the features from this tutorial: https://sklearn-crfsuite.readthedocs.io/en/latest/tutorial.html#features.
But I did not understand the process of making the model. After doing the training, I checked the output on the features and got something like this
As you can see there are lots of numbers. I know these numbers will be used as a reference to take the probability of doing entity recognition. But how? I need to know the process to get the numbers. How can CRF do entity recognition using probability?