KWiecko

For a single record data frame train_test_split() sometimes assigns this single record to test set.

4

**Describe the issue**: _Disclaimer: I know the bug looks silly but I still wanted to give a heads up._ For a single data frame with only 1 record `train_test_split()` sometimes...

"The truth value of a Series is ambiguous" exception when training with `sample_weights`.

2

**What happened**: When `sample_weight` is specified and train() method input's npartitions is not equal to the number of workers the following evaluation: sample_weight = concat(sample_weight) if np.all(sample_weight) else None fails...

Work stealing causes a sudden death of workers without any informative traceback.

6

**Describe the issue**: I'm sorry I can't share the code (it is confidential) but I'll try to describe the 'crash' scenario as best I can. I'll start with brief data...