[Solved] ConvergenceWarning: lbfgs failed to converge (status=1): STOP: TOTAL NO. of ITERATIONS REACHED L

Last Updated on 2021-10-30 by Clay

The full error message as follows:

Start to train the model.
\sklearn\linear_model\_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):

STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.

Increase the number of iterations (max_iter) or scale the data as shown in:
    https://scikit-learn.org/stable/modules/preprocessing.html
Please also refer to the documentation for alternative solver options:
    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression  n_iter_i = _check_optimize_result(

\sklearn\linear_model\_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):

STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.

Increase the number of iterations (max_iter) or scale the data as shown in:
    https://scikit-learn.org/stable/modules/preprocessing.html
Please also refer to the documentation for alternative solver options:
    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression  n_iter_i = _check_optimize_result(

The meaning of the error message is lbfgs cannot converge because the iteration number is limited and aborted.

---

Solution

There are three solutions:

• Increase the iterable number (max_iter default is 100)
• Reduce the data scale
• Change the solver

---

References

• https://stackoverflow.com/questions/62658215/convergencewarning-lbfgs-failed-to-converge-status-1-stop-total-no-of-iter
• https://stats.stackexchange.com/questions/184017/how-to-fix-non-convergence-in-logisticregressioncv
• https://github.com/scikit-learn/scikit-learn/issues/10866

---

Read More

• [Scikit-Learn] Using “train_test_split()” to split your data
• [Scikit-Learn] Tutorial (0) What is “Scikit-Learn”