'Why does my model exactly follow the trend of only one input out of several?
I am a student studying deep learning. This time, I am making a deep learning model in a certain project. This model is a regression model that takes several properties as input. After checking the model's training and testing values, I found that this model only follows the trend of one particular input attribute. Of course, not every time, but most of the time. I analyzed and calculated the correlation value between each input value and the label value. This input has a significantly higher correlation with the label compared to other input values.
- Is this case normal?
- Is it the right way to check by doing a correlation calculation(Each Input-Label)?
- How can I reduce the impact of certain input properties?
I apologize for asking these questions due to my poor English skills and insufficient study. Thank you.
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|