Is it possible, for any Python sklearn learning model, to update an already-fitted model? For example, if I have trained my model on a large set of data, is it possible to update the training by introducing new data, or does it have to be retrained every time?
Asked
Active
Viewed 23 times
0

desertnaut
- 57,590
- 26
- 140
- 166

Falcon72
- 1
- 2
-
2Most models have to be trained from scratch. Some models have `partial_fit()` method meaning you can update it – Sergey Bushmanov Oct 22 '20 at 18:09
-
See the duplicate thread and the [Incremental learning](https://scikit-learn.org/stable/modules/computing.html#incremental-learning) section in the docs. – desertnaut Oct 22 '20 at 20:55