test: cover negative mixture fractions in any position

Signed-off-by: pengmin <minpeng@semigraph.com>

Author: pengmin

tests/test_cli_utils.py

@@ -390,6 +390,7 @@ def test_dataset_fraction_partial_raises_error(self):
         [
             ([0.0, 0.0], "must be greater than zero"),
             ([-0.5, 1.5], "must be non-negative"),
+            ([1.5, -0.5], "must be non-negative"),
         ],
     )
     def test_dataset_fraction_invalid_raises_error_before_loading(self, fractions, error_message):

trl/scripts/utils.py

@@ -65,7 +65,7 @@ class DatasetConfig:
             datasets, and the mixture size is capped so that no dataset is oversampled: the first `round(fraction * N)`
             rows of each dataset are kept, where `N` is the largest mixture size that avoids oversampling. Must be set
             for either all datasets in the mixture or none of them. Not supported for streaming datasets. When unset,
-            the full datasets are concatenated.
+            the full datasets are concatenated. Values must be non-negative and their sum must be greater than zero.
     """
 
     path: str