Scholarlink

This research work is based on how well gaming systems performance affects the frames per second (FPS) of the game. We tried to predict the FPS with the help of ensemble learning or mainly stacking which is a part of ensemble learning. It is also found from this study that FPS mainly does not depend on the GPU entirely but also depends on the CPU. The data set is obtained from Open ML which was posted a year back where it contains the CPU specifications with the FPS that they get. So, we used this data and an ensembling approach so that we can predict accurate predictions of fps. We got an r2 value of 0.99 which is a very good r2 score and is close to 1. So, we can say that this model can be used for benchmarking systems. We used a Catboost and histogram gradient boosting techniques as a base estimator with linear regression as the meta-estimator which gave this higher performance in our case. This approach is very resilient and non-time-consuming so it can be used on any website to get input and get an output of FPS.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prediction of Fps Using Ensembling Approach for Benchmarking Gaming Systems

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • D. Ramkumar,
  • K. Jeeveth,
  • Pooja N. Umekar,
  • Jyoti C. Bambal,
  • Sagar Dhanraj Pande,
  • Anil V. Turukmane

摘要

This research work is based on how well gaming systems performance affects the frames per second (FPS) of the game. We tried to predict the FPS with the help of ensemble learning or mainly stacking which is a part of ensemble learning. It is also found from this study that FPS mainly does not depend on the GPU entirely but also depends on the CPU. The data set is obtained from Open ML which was posted a year back where it contains the CPU specifications with the FPS that they get. So, we used this data and an ensembling approach so that we can predict accurate predictions of fps. We got an r2 value of 0.99 which is a very good r2 score and is close to 1. So, we can say that this model can be used for benchmarking systems. We used a Catboost and histogram gradient boosting techniques as a base estimator with linear regression as the meta-estimator which gave this higher performance in our case. This approach is very resilient and non-time-consuming so it can be used on any website to get input and get an output of FPS.