Cannot save HardSamplingClassifier

[nehalAggarwal]

Hello,

I have run models using HardSamplingClassifier and I wished to save the model using pickle or joblib.

It seems that the model is not ready to be saved as it gives Pickling Error as -

Is there a way I can fix this error and save/pickle HardSamplingClassifier model?

[raphaelsty]

Hi @nehalAggarwal,

Pickle does not allow serializing lambda function. The HardSampling class is one of the few classes in River to have a lambda function in it's source code. I suggest you to use the Dill library to serialize this model. Dill is more or less the same as Pickle, maybe a bit slower but allows to serialize lambda functions.

You can install Dill using:

pip install dill

from river import datasets
from river import evaluate
from river import imblearn
from river import linear_model
from river import metrics
from river import optim
from river import preprocessing

model = (
 preprocessing.StandardScaler() |
 imblearn.HardSamplingClassifier(
     classifier=linear_model.LogisticRegression(),
     p=0.1,
     size=40,
     seed=42,
 )
)

import dill

with open('./model.pkl', 'wb') as output_file:
    dill.dump(model, output_file)

import dill

with open('./model.pkl', 'rb') as input_file:
    model = dill.load(input_file)

[MaxHalford]

@raphaelsty is right, and yet we could and should fix HardSamplingClassifier so that it doesn't use a lambda anymore.

[MaxHalford]

This is fixed in #545.