Python Forum

The following code generates an error when I try to interpret it with python.

svm_c = GridSearchCV(
    cv=TimeSeriesSplit(gap=0, max_train_size=None, n_splits=5, test_size=None),
    estimator=SGDClassifier(n_jobs=-1, random_stat=21),
    n_jobs=-1,
    param_grid={"alpha": [0.0001, 0.001, 0.01, 0.1, 1, 10, 100, 1000]},
    return_train_score=True,
    scoring="f1_macro",
)

The error is:

TypeError                                 Traceback (most recent call last)
Input In [187], in <cell line: 1>()
      1 svm_c = GridSearchCV(
      2     cv=TimeSeriesSplit(gap=0, max_train_size=None, n_splits=5, test_size=None),
----> 3     estimator=SGDClassifier(n_jobs=-1, random_stat=21),
      4     n_jobs=-1,
      5     param_grid={"alpha": [0.0001, 0.001, 0.01, 0.1, 1, 10, 100, 1000]},
      6     return_train_score=True,
      7     scoring="f1_macro",
      8 )

TypeError: SGDClassifier.__init__() got an unexpected keyword argument 'random_stat'

I am not sure why. I googled the error, and it claimed that scikit-learn was the problem. I have version 1.1.1. It says that
I need version 0.18. I upgraded scikit-learn to 1.1.2 and got the same error when I tried to interpret it.

I should emphasize that I am only running this code snippet. It is part of a larger program, but I am just working it this piece now.

I cannot upgrade scikit-learn to version 0.18. That is impossible.

I used the command

pip install scikit-learn==0.18 --force upgrade 

and all it did was give a lot of errors. Why is this error that I discussed above happening? I am using
python 3.10.4.

I suspect it has to do with a specific version of scikit-learn.


Respectfully,

Led_Zepplin

SGDClassifier expects random_state but you gave random_stat