I might have missed this in the documentation - what are the minimum number of samples for control and treatment required for the analysis?
Follow up: error disappeared after including more samples. Would be helpful to add the minimum number of samples required in the documentation
Traceback (most recent call last):
File "/Users/y1weng/Downloads/20231129_qadabra_gutb/mi04t1_qadabra_out/.snakemake/scripts/tmp9w0fsvnp.logistic_regression.py", line 54, in <module>
folds = [(train, test) for train, test in cv.split(X, y)]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/y1weng/Downloads/20231129_qadabra_gutb/mi04t1_qadabra_out/.snakemake/scripts/tmp9w0fsvnp.logistic_regression.py", line 54, in <listcomp>
folds = [(train, test) for train, test in cv.split(X, y)]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/y1weng/Downloads/20231129_qadabra_gutb/mi04t1_qadabra_out/.snakemake/conda/ae102f57fe1fb6ba91271bbd0ace3284_/lib/python3.11/site-packages/sklearn/model_selection/_split.py", line 1527, in split
for train_index, test_index in cv.split(X, y, groups):
File "/Users/y1weng/Downloads/20231129_qadabra_gutb/mi04t1_qadabra_out/.snakemake/conda/ae102f57fe1fb6ba91271bbd0ace3284_/lib/python3.11/site-packages/sklearn/model_selection/_split.py", line 377, in split
for train, test in super().split(X, y, groups):
File "/Users/y1weng/Downloads/20231129_qadabra_gutb/mi04t1_qadabra_out/.snakemake/conda/ae102f57fe1fb6ba91271bbd0ace3284_/lib/python3.11/site-packages/sklearn/model_selection/_split.py", line 108, in split
for test_index in self._iter_test_masks(X, y, groups):
File "/Users/y1weng/Downloads/20231129_qadabra_gutb/mi04t1_qadabra_out/.snakemake/conda/ae102f57fe1fb6ba91271bbd0ace3284_/lib/python3.11/site-packages/sklearn/model_selection/_split.py", line 770, in _iter_test_masks
test_folds = self._make_test_folds(X, y)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/y1weng/Downloads/20231129_qadabra_gutb/mi04t1_qadabra_out/.snakemake/conda/ae102f57fe1fb6ba91271bbd0ace3284_/lib/python3.11/site-packages/sklearn/model_selection/_split.py", line 732, in _make_test_folds
raise ValueError(
ValueError: n_splits=5 cannot be greater than the number of members in each class.
I might have missed this in the documentation - what are the minimum number of samples for control and treatment required for the analysis?
Follow up: error disappeared after including more samples. Would be helpful to add the minimum number of samples required in the documentation