diff --git a/pytorch_tabnet/multitask.py b/pytorch_tabnet/multitask.py
index 9666bf4f..c69d4cdd 100644
--- a/pytorch_tabnet/multitask.py
+++ b/pytorch_tabnet/multitask.py
@@ -3,7 +3,7 @@
 from scipy.special import softmax
 from pytorch_tabnet.utils import PredictDataset, filter_weights
 from pytorch_tabnet.abstract_model import TabModel
-from pytorch_tabnet.multiclass_utils import infer_multitask_output
+from pytorch_tabnet.multiclass_utils import infer_multitask_output, check_output_dim
 from torch.utils.data import DataLoader
 
 
@@ -60,6 +60,9 @@ def stack_batches(self, list_y_true, list_y_score):
 
     def update_fit_params(self, X_train, y_train, eval_set, weights):
         output_dim, train_labels = infer_multitask_output(y_train)
+        for _, y in eval_set:
+            for task_idx in range(y.shape[1]):
+                check_output_dim(train_labels[task_idx], y[:, task_idx])
         self.output_dim = output_dim
         self.classes_ = train_labels
         self.target_mapper = [