Forecasting of GPU Prices Using Transformer Method
Abstract
Abstract— GPU or VGA (graphic processing unit) is a vital component of computers and laptops, used for tasks such as rendering videos, creating game environments, and compiling large amounts of code. The price of GPU/VGA has fluctuated significantly since the start of the COVID19 pandemic in 2020. This research aims to forecast future GPU prices using deep learning-based time series forecasting using the Transformer model. We use daily prices of NVIDIA RTX 3090 Founder Edition as a test case. We use historical GPU prices to forecast 8, 16, and 30 days. Moreover, we compare the results of the Transformer model with two other models, RNN and LSTM. We found that to forecast 30 days; the Transformer model gets a higher coefficient of correlation (CC) of 0.8743, a lower root mean squared error (RMSE) value of 34.68, and a lower mean absolute percentage error (MAPE) of 0.82 compared to the RNN and LSTM model. These results suggest that the Transformer model is an effective and efficient method for predicting GPU prices.
Keywords— GPU, Transformer, Forecasting, Time Series Forecasting
References
REFRENCES
The Economist,
blame for the graphics-chip shortage,= 2021.
https://www.usnews.com/news/top-news/articles/2022-09-
/nvidia-unveils-new-gaming-chip-with-ai-features-tapstsmc-formanufacturing#:~:text=Nvidia%20designs%20its%20chips
%20but,by%20Samsung%20Electronics%20Co%20Ltd.
(accessed Apr. 25, 2022).
Z. Zhao et al.,
on the Transformer Model,= Information (Switzerland), vol.
, no. 12, Dec. 2021, doi: 10.3390/INFO12120516.
S. A. A. Leksono, Z. G. Prastyawan, and I.
Veriawati,
oleh Nilai Bitcoin,= JURNAL ILMIAH FIFO, vol. XI, no. 1,
pp. 65-74, Apr. 2019.
M. Chlebus, M. Dyczko, and M. Wozniak,
Techniques for Time Series Forecasting Problem,= Central
European Economic Journal, vol. 8, no. 55, pp. 44-62, Jan.
, doi: 10.2478/ceej-2021-0004.
Maxime,
https://medium.com/inside-machine-learning/what-is-atransformer-d07dd1fbec04 (accessed Mar. 09, 2022).
Transformer Vs RNN in Speech Applications,= ASRU, 2019.
[Online]. Available: http://www.merl.com
A. Zeyer, P. Bahar, K. Irie, R. Schluter, and H. Ney,
Models for ASR,= in 2019 IEEE Automatic Speech
Recognition and Understanding Workshop, ASRU 2019 -
Proceedings, Dec. 2019, pp. 8-15. doi:
1109/ASRU46091.2019.9004025.
G. A. Galindo Padilha, J. R. Ko, J. J. Jung, and P. S.
G. de Mattos Neto,
Model for Multivariate Renewable Energy,= Applied
Sciences (Switzerland), vol. 12, no. 21, Nov. 2022, doi:
3390/app122110985.
S. Li et al.,
the Memory Bottleneck of Transformer on Time Series
Forecasting,= Jun. 2019.
N. Wu, B. Green, X. Ben, and S. O'Banion,
Transformer Models for Time Series Forecasting: The
Influenza Prevalence Case,= Jan. 2020, [Online]. Available:
http://arxiv.org/abs/2001.08317
A. Vaswani et al.,
Han'guk T'ongsin Hakhoe, IEEE Communications
Society, Denshi J__h__ Ts__shin Gakkai (Japan). Ts__shin
Sosaieti, and Institute of Electrical and Electronics Engineers,
RNN-based Deep Learning for One-hour ahead Load
Forecasting. 2020.
H. Apaydin, H. Feizi, M. T. Sattari, M. S. Colak, S.
Shamshirband, and K. W. Chau,
recurrent neural network architectures for reservoir inflow
forecasting,= Water (Switzerland), vol. 12, no. 5, May 2020,
doi: 10.3390/w12051500.
D. Zhang, Q. Peng, J. Lin, D. Wang, X. Liu, and J.
Zhuang,
neural network algorithm,= Water (Switzerland), vol. 11, no.
, Apr. 2019, doi: 10.3390/w11040865.
M. S. Hossain and H. Mahmood,
photovoltaic power forecasting using an LSTM neural
network and synthetic weather forecast,= IEEE Access, vol.
, pp. 172524-172533, 2020, doi:
1109/ACCESS.2020.3024901.
S. R. Venna, A. Tavanaei, R. N. Gottumukkala, V.
v. Raghavan, A. S. Maida, and S. Nichols,
Access, vol. 7, pp. 7691-7701, 2019, doi:
1109/ACCESS.2018.2888585.
P. Schober and L. A. Schwarte,
coefficients: Appropriate use and interpretation,= Anesth
Analg, vol. 126, no. 5, pp. 1763-1768, May 2018, doi:
1213/ANE.0000000000002864.
M. A. Istiake Sunny, M. M. S. Maswood, and A. G.
Alharbi,
LSTM and Bi-Directional LSTM Model,= in 2nd Novel
Intelligent and Leading Emerging Sciences Conference,
NILES 2020, Oct. 2020, pp. 87-92. doi:
1109/NILES50944.2020.9257950.
A. de Myttenaere, B. Golden, B. le Grand, and F.
Rossi,
models,= Neurocomputing, vol. 192, pp. 38-48, Jun. 2016,
doi: 10.1016/j.neucom.2015.12.114.
Keepa,
Edition Graphics Card,= 2019.
https://keepa.com/#!product/1-B08HR6ZBYJ (accessed
May 04, 2022).
