ValueError: Error: Input data is not in a valid format

I have the following code:
```
tpot_dict[i].fit(train_dfs[i], ys_dfs[i])
print(tpot_dict[i].score(train_dfs[i].as_matrix(), ys_dfs[i].as_matrix()))
name = 'segment ' + str(i) + '.py'
tpot_dict[i].export(name)
```
Fiting several models defined in a dictionary. I run for "n" generations.
The problem is that at the last "N" generation, the program crashes, showing the error:
```
1077 'Error: Input data is not in a valid format. Please confirm '
1078 'that the input data is scikit-learn compatible. For example, '
1079 'the features must be a 2-D array and target labels must be a '

ValueError: Error: Input data is not in a valid format. Please confirm that the input data is scikit-learn compatible. For example, the features must be a 2-D array and target labels must be a 1-D array.
```
I have run this w/o the .as_matrix() part, with same results.
I have run then w/o the dictionary part, with same results.
My DFs does not have any "NA" or non numerical data. The problem is strange becuase it is always happening in the last generation.

I do have the following:

```
TPOTRegressor(generations=25, scoring = 'neg_mean_absolute_error',
cv=5, random_state=65,
periodic_checkpoint_folder = 'tpot',
verbosity=2,
early_stop=5,
n_jobs=-1)
```
Any feedback?