[R-package] ensure boosting happens in tests on small datasets

[jameslamb]

The R package contains a few tests using the mtcars dataset that comes built into R.

That dataset only has 32 observations in it. From ?mtcars.

data(mtcars)
dim(mtcars)
# [1] 32 11

As a result, the calls to lightgbm() and lgb.train() in tests using that dataset are not currently performing any boosting, for reasons described in #5081.

If setting verbosity to something low enough to allow INFO and WARNING level logs, those tests contain logs like the following:

[LightGBM] [Warning] There are no meaningful features which satisfy the provided configuration. Decreasing Dataset parameters min_data_in_bin or min_data_in_leaf and re-constructing Dataset might resolve this warning.
[LightGBM] [Info] Total Bins 0
[LightGBM] [Info] Number of data points in the train set: 32, number of used features: 0
[LightGBM] [Info] Start training from score 20.090625
[LightGBM] [Warning] Stopped training because there are no more leaves that meet the split requirements
[LightGBM] [Warning] Stopped training because there are no more leaves that meet the split requirements
[LightGBM] [Warning] Stopped training because there are no more leaves that meet the split requirements

This PR proposes setting min_data_in_bin = 1 and min_data_in_leaf = 1 in those examples, to ensure that boosting occurs. I believe this will improve the test coverage LightGBM gets from these tests, by ensuring that the tests use models that actually generate trees with splits.

[StrikerRUS]

LGTM, thanks!

[github-actions]

This pull request has been automatically locked since there has not been any recent activity since it was closed. To start a new related discussion, open a new issue at https://github.com/microsoft/LightGBM/issues including a reference to this.