Authors

Le Duc Tinh 1 , Huynh Nguyen Dinh Quoc 2 , Nguyen Gia Trong 1,3*

Affiliations

 1 Faculty of Geomatics and Land Administration, Hanoi University of Mining and Geology; leductinh@humg.edu.vn; nguyengiatrong@humg.edu.vn

2 Ho Chi Minh City University of Natural Resources and Environment; hndquoc@hcmunre.edu.vn

3 Geodesy and Environment research group, Hanoi University of Mining and Geology; nguyengiatrong@humg.edu.vn

*Corresponding author: nguyengiatrong@humg.edu.vn; Tel.: +84–963124980 

Abstracts

This study applies the GRU (Gated Recurrent Unit) model when selecting different values of batch-size, namely 16, 32, and 64, with varying epochs of 20, 50, 100, 150, and 200. The input data comprises observations collected by two GNSS CORS stations from the VNGEONET network, namely HYEN and CTHO, spanning from August 10, 2019, to March 18, 2022. Initially, GNSS CORS data is processed using Gamit/Globk software to obtain the Up-component, which serves as the input data for the GRU model. The research results indicate that the statistical performance metrics of the model, such as RMSE and MAE, decrease while the F-Score increases when the batch-size decreases and the epoch value increases. In cases where the Up-component exhibits irregular variations (seasonal fluctuations), the performance of the GRU model is subpar, with an F-Score of 0 observed when batch-size values are 32 and 64 and epoch value is 20. For data following the pattern of CTHO CORS station, the GRU model performs exceptionally well when batch-size is 16 and epoch is 200. However, the forecasting performance is low for data from HYEN CORS station, indicating the need for further investigation in the future.

Keywords

Artificial Intelligence; Batch size; Epoch; GRU; GNSS time-series.

Cite this paper

Tinh, L.D.; Quoc, H.N.D.; Trong, N.G. Exploring the training results of machine learning models using different batch sizes and epochs: A case study with GNSS time series data. J. Hydro-Meteorol. 2024, 19, 90-99.

References

1. Yu, K.; Rizos C.; Burage, D.; Dempster, A.G.; Zhang, K.; Markgraf, M. An overview of GNSS remote sensing. EURASIP J. Adv. Signal Process. 2014, 134, 1–14.

2. Bastos, L.; Bos, M.; Fernandes, R.M. Deformation and tectonics: contribution of GPS measurements to plate tectonics–overview and recent developments. Sci. Geodesy-I. 2010, 155–184.

3. Srinivasan, M.; Tsontos, V. Satellite altimetry for ocean and coastal applications: A review. Remote Sens. 2023, 15(16), 3939. https://doi.org/10.3390/rs15163939.

4. Gómez, C.; White, J.C.; Wulder, M.A. Optical remotely sensed time series data for land cover classification: A review. ISPRS J. Photogramm. Remote Sens. 2016, 116, 55–72.

5. Banskota, A.; Kayastha N.; Falkowski M.J.; Wulder M.A.; Froese R.E.; White J.C. Forest monitoring using Landsat time series data: A review. Can. J. Remote Sens. 2014, 40(5), 362–384.

6. Vrieling, A.J.C. Satellite remote sensing for water erosion assessment: A review. Catena 2006, 65(1), 2–18.

7. Li, S.; Dragicevic, S.; Castro, F.A.; Sester, M.; Winter, S.; Coltekin, A.; Pettit, A.; Jiang, B.; Haworth, J.; Stein, A.; Cheng, T. Geospatial big data handling theory and methods: A review and research challenges. ISPPS J. Photogramm. Remote Sens.  2016, 115, 119–133.

8. Trong, N.G.; Tinh, L.Đ.; Cuong, N.V.; Quang, P.N. GNSS Data Processing: Theory, Software, and Applications. Transport Publishing House, 2023, pp. 300.

9. Li, Y. Analysis of GAMIT/GLOBK in high-precision GNSS data processing for crustal deformation. Earthquake Res. Adv. 2021, 1(8-11), 100028. https://doi.org/10.1016/j.eqrea.2021.100028.

10. Cetin, S.; Aydin, C.; Dogan, U. Comparing GPS positioning errors derived from GAMIT/GLOBK and Bernese GNSS software packages: A case study in CORS-TR in Turkey. Surv. Rev. 2019, 51(369), 533–543.

11. Klos, A.; Bogusz, J.; Bos, M.S.; Gruszczynska, M. Modelling the GNSS time series: different approaches to extract seasonal signals. Geodetic Time Ser. Anal. Earth Sci.  2020, pp. 211–237.

12. Wang, J.; Nie, G.; Gao, S.; Wu, S.; Li, H.; Ren, X. Landslide deformation prediction based on a GNSS time series analysis and recurrent neural network model. Remote Sens. 2021, 13(6), 1055. https://doi.org/10.3390/rs13061055.

13. Gao, W.; Li, Z.; Chen, Q.; Jiang, W.; Feng, Y. Modelling and prediction of GNSS time series using GBDT, LSTM and SVM machine learning approaches. J. Geod. 2022, 96(10), 71. https://doi.org/10.1007/s00190-022-01662-5.

14. Li, Z.; Lu, T.; Yu, K.; Wang, J. Interpolation of GNSS position time series using GBDT, XGBoost, and RF machine learning algorithms and models error analysis. Remote Sens. 2023, 15(18), 4374. https://doi.org/10.3390/rs15184374.

15. Hoffer, E.; Hubara, I.; Soudry, D. Train longer, generalize better: Closing the generalization gap in large batch training of neural networks. Adv. Neural Inf. Process. Syst. 2017, 1–13.

16. McCandlish, S.; Kaplan, J.; Amodei, D.; Team, O.D. An empirical model of large-batch training. ArXiv 2018, 1–35. https://doi.org/10.48550/arXiv.1812.06162.

17. Masters, D.; Luschi, C. Revisiting small batch training for deep neural networks. Comput. Sci. Mach. Learn. 2018, 1–18. https://doi.org/10.48550/arXiv.1804.07612.

18. Thai, T.H.; Khiem, M.V.; Thuy, N.B.; Ha, B.M.; Ngoc, P.K. Building a regression neural network model to predict significant wave heights at Con Co station, Quang Tri, Vietnam. J. Hydro-Meteorol. 2022, EME4, 73–84.

19. Phong, D.V.; Trong, N.G.; Chien, N.V.; Thanh, N.H.; Ha, L.L.; Quan, N.V.; Quang, P.N. Analysis of land vertical movement using ANN function from the results of processing GNSS time series data. J. Hydro-Meteorol. 2023, 752, 41–50.

20. Phong, N.D.; Duong, H.H. Application of deep learning models in forecasting surface water quantity of Bac Hung Hai irrigation system. Water Resour. Mag. 2023, 1, 61–72.

21. Available online: https://geoweb.mit.edu/gg/.

22. Trong, N.G.; Nghia, N.V.; Khai, P.C.; Thanh, N.H.; Ha, L.L.; Dung, V.T.; Quan, N.V.; Quang, P.N. Determination of tectonic velocities in Vietnam territory based on data of CORS stations of VNGEONET network. J. Hydro-Meteorol. 2022, 739, 59–66.

23. Available online: https://www.python.org/.

24. Available online: https://anaconda.org/anaconda/pandas.

25. Savchuk, S.; Doskich, S.; Golda, P.; Rurak, A. The Seasonal Variations Analysis of Permanent GNSS Station Time Series in the Central-East of Europe. Remote Sens. 2023, 15(15), 3858. https://doi.org/10.3390/rs15153858.

26. Carbonari, R.; Riccardi, U.; Martino, P.D.; Cecere, G.; Maio, R.D. Wavelet-like denoising of GNSS data through machine learning. Application to the time series of the Campi Flegrei volcanic area (Southern Italy). Geomatics Nat. Hazards Risk 2023, 14(1), 2187271. https://doi.org/10.1080/19475705.2023.2187271.