# Dataset handling and model selection

**URL:** <https://discuss.pytorch.org/t/dataset-handling-and-model-selection/218106>\
**Category:** Uncategorized\
**Created:** [March 21, 2025, 6:12am UTC](https://discuss.pytorch.org/t/dataset-handling-and-model-selection/218106 "2025-03-21T06:12:00Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Muhammad\_Irfan](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/muhammad_irfan/32/76272_2.png) [@Muhammad\_Irfan](https://discuss.pytorch.org/u/Muhammad_Irfan)\
**Post date:** [March 21, 2025, 6:12am UTC](https://discuss.pytorch.org/t/dataset-handling-and-model-selection/218106/1 "2025-03-21T06:12:00Z")

</div>

I am facing a problem, i have a point error dataset of 74 rows in which 3 are my inputs and 16 are output features. Well when I trained my model using polynomial regression, if I increased the degree I got good results in train r2 and test r2 and this is overfitting. when I used DNN model with complex hidden layer I got good train r2 but test r2 fails so what you suggest to get good train and test r2, either I need to change my model as when I tuned my polynomial model the results not improved or we have to collect more data ?