# Most Influential Variables?

**URL:** <https://discuss.pytorch.org/t/most-influential-variables/185209>\
**Category:** Uncategorized\
**Created:** [July 27, 2023, 8:10pm UTC](https://discuss.pytorch.org/t/most-influential-variables/185209 "2023-07-27T20:10:00Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![aeroboyboiler](https://discuss.pytorch.org/letter_avatar_proxy/v4/letter/a/e5b9ba/32.png) [@aeroboyboiler](https://discuss.pytorch.org/u/aeroboyboiler)\
**Post date:** [July 27, 2023, 8:10pm UTC](https://discuss.pytorch.org/t/most-influential-variables/185209/1 "2023-07-27T20:10:00Z")

</div>

My team is doing quite a simple regression model using PyTorch. We would like to determine what variables have the most “influence” on the curve fit. Basically what variables are driving the regression model and what are the percentages of influence. Does PyTorch have this capability?