Making a ML model scikit-learn compatible

I want to make this ML model scikit-learn compatible: https://github.com/manifoldai/merf

To do that, I followed the instructions here: https://danielhnyk.cz/creating-your-own-estimator-scikit-learn/ and imported from sklearn.base import BaseEstimator, RegressorMixin and inherited from them like so: class MERF(BaseEstimator, RegressorMixin):

However, when I check for scikit-learn compatibility:

from sklearn.utils.estimator_checks import check_estimator

import merf
check_estimator(merf)

I get this error:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "C:\Users\hap\anaconda3\envs\a1\lib\site-packages\sklearn\utils\estimator_checks.py", line 500, in check_estimator
    for estimator, check in checks_generator:
  File "C:\Users\hap\anaconda3\envs\a1\lib\site-packages\sklearn\utils\estimator_checks.py", line 340, in _generate_instance_checks
    yield from ((estimator, partial(check, name))
  File "C:\Users\hap\anaconda3\envs\a1\lib\site-packages\sklearn\utils\estimator_checks.py", line 340, in <genexpr>
    yield from ((estimator, partial(check, name))
  File "C:\Users\hap\anaconda3\envs\a1\lib\site-packages\sklearn\utils\estimator_checks.py", line 232, in _yield_all_checks
    tags = estimator._get_tags()
AttributeError: module 'merf' has no attribute '_get_tags'

How do I make this model scikit-learn compatible?

from Making a ML model scikit-learn compatible