bluesheeptoken / subseq Goto Github PK

One example input of the predictor seems to be model.fit([["A", "B", "C"], ["A", "C"], ["A", "B"]]) , here "A", "B", "C" can be products purchased by three customers and each sublist is a ordered sequence of each customer's purchase history. But your model seems unable to deal with model.fit([["A", {"B", "C"}, "D"], ["A", {"A", "B"} "C"], ["A", "B"]]) where each set is a basket of products that are purchased at the same timestamp. How can I do sequence prediction with the input data that have the form like this: http://philippe-fournier-viger.com/spmf/CPTPlus.php

great code thanks

but do you have example that your code is better than neural networks based code ?

Hi,
Thanks for this package. I was wondering if there was an option to access the top k predictions or least get the probabilities.
thanks !

Hi! Is there any way I can see the probability? Something like:

model = Subseq(1)
model.probability()

Hi! In fedora 38 and Ubuntu 23.04, an error occurs when installing the package
Python 3.11.2

P.s. It looks like the problem is due to Python 3.11, in Python 3.10 on the same operating systems, the installation goes without problems.

Hi!
First of all thank you for this package - model produces great results.
But we faced with problem that the model can't be saved. We can't neither pickle nor dill it as it has non-serializable object 'thisptr'.
Example:

from dill import dumps
from subseq.subseq import Subseq

model = Subseq(1)
model.fit([['hello', 'world']])
dumps(model)

execs with exception

  File "<input>", line 1, in <module>
  File "\lib\site-packages\dill\_dill.py", line 304, in dumps
    dump(obj, file, protocol, byref, fmode, recurse, **kwds)#, strictio)
  File "\lib\site-packages\dill\_dill.py", line 276, in dump
    Pickler(file, protocol, **_kwds).dump(obj)
  File "\lib\site-packages\dill\_dill.py", line 498, in dump
    StockPickler.dump(self, obj)
  File "\lib\pickle.py", line 437, in dump
    self.save(obj)
  File "\lib\pickle.py", line 524, in save
    rv = reduce(self.proto)
  File "stringsource", line 2, in subseq.subseq.Subseq.__reduce_cython__
TypeError: self.thisptr cannot be converted to a Python object for pickling

Do you have any solution how to store the resulting model in storage?