i got a problem when i use Logistic Regression (Learner). Die Error is:
ERROR Logistic Regression (Learner) Execute failed: The dataset must have at least 1125 rows, but it has only 349 rows. It is recommended to use a larger dataset in order to increase accuracy.
But after i raised the rows of the dataset, i got a nother error like this:
ERROR Logistic Regression (Learner) Execute failed: The dataset must have at least 2375 rows, but it has only 1166 rows. It is recommended to use a larger dataset in order to increase accuracy.
Can everyone tell me what is the problem? It seems like not the Problem in the nummer of rows. Thanks a lot!
At a first glance, it seems strange that raising the number of rows raises the number of needed rows, too.
But in fact it can happen when either the target column or one of the inlcuded columns are of type String (moninal data). The number of rows that the logistic regression solver needs, depends on the number of distinct values (categories) in these columns.
I assume that the new rows, added categories to your nominal data columns.
I hope this helps.
How many columns does your dataset contain as the number of rows need to be significantly greater than the number of columns. You may find it easier to be more selective with the columns you use (the independent variables).
yes, the new rows added categories to the nominal data columns. Now the problem is solved by targeted selection. Thanks a lot!