Root-mean-square error (RMSE) or mean absolute error (MAE): when to use them or not

Hodson, Timothy O.

doi:10.5194/gmd-15-5481-2022

Cited by 576 publications

(213 citation statements)

References 24 publications

Supporting

Mentioning

203

Contrasting

Unclassified

Order By: Relevance

“…Novelty, the means of these metrics could be reference here [14]. For the regression task, we performed the means root-mean-square error (RMSE) metrics [15]. For the classification task, we computed the area under the receiver operating characteristic curve/precision recall curve (AUC/PRC) metrics [16].…”

Section: Methodsmentioning

confidence: 99%

iupacGPT: IUPAC-based large-scale molecular pre-trained model for property prediction and molecule generation

Mao

Wang

Cho

et al. 2023

Preprint

View full text Add to dashboard Cite

The IUPAC (International Union of Pure and Applied Chemistry) nomenclature is a globally recognized unique naming system which assigns names to chemical compounds. As a form of molecular representation closest to natural language, it allows to estimate molecular data in a large-scale pre-trained paradigm by employing machine learning approaches for natural language processing (NLP). Although, SMILES is currently popular molecular representation used by most generative models, different molecular representation is suitable for different scenarios, and considering the advantages of IUPAC in terms of readability, it becomes meaningful to explore the difference of these two different molecular representations for molecular generation and regression/classification tasks. In this paper, we attempt to adapt the capabilities of transformer to a large IUPAC corpus by constructing a GPT-2-like language model named iupacGPT. For each task in addition to the molecular generation, we freeze model parameters and attach trainable lightweight networks to fine tune. The results show that pre-trained iupacGPT can capture general knowledge that can be successfully transferred to the downstream tasks such as molecule generation and binary classification and property regression prediction. What’s more, with a same setup, iupacGPT outperforms the model smilesGPT in term of the downstream tasks. Overall, transformer-like language models pretrained on IUPAC corpora are promising alternatives that obtain more intuitive in terms of interpretability and semantic level than on SMILES corpora, and scale well with the pretraining data size. https://github.com/AspirinCode/iupacGPT

show abstract

Section: Methodsmentioning

confidence: 99%

iupacGPT: IUPAC-based large-scale molecular pre-trained model for property prediction and molecule generation

Mao

Wang

Cho

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Two key metrics are used to evaluate the performance of the model proposed in this study, namely, the root mean square error (RMSE) and the coefficient of determination ( R 2 ). , RMSE is a widely used index that quantifies the discrepancy between the actual and predicted values of the response variable. It is calculated using the following equation: RMSE = 0.25em 1 K 0.25em ∑ i = 1 K ( y false( i false) − ŷ false( i false) ) 2 where y ( i ) represents the experimental value, ŷ ( i ) is the estimated value, and K is the number of data points. The coefficient of determination ( R 2 ) measures the degree of fit of a regression model to the data, with values ranging from 0 to 1 .…”

Section: Modeling and Optimization Of Fabrication Strategies For Pseu...mentioning

confidence: 99%

Optimizing the Ultrashort Laser Pulses for In Situ Nanostructure Generation Technique for High-Performance Supercapacitor Electrodes Using Artificial Neural Networks and Simulated Annealing Algorithms

Khosravinia

Kiani

2023

ACS Omega

View full text Add to dashboard Cite

Transition metals (TMs) are being investigated as electrodes for pseudocapacitors, where an oxide layer is necessary to allow for rapid redox reactions. In this work, we utilized an in situ, rapid, binder-free, and green method for the fast fabrication of pseudocapacitor electrodes called ultrashort laser pulses for in situ nanostructure generation (ULPING) to form oxide layers on a titanium sheet. By utilizing this fabrication technique on a titanium sheet, a specific areal capacitance of 0.3579 mF cm −2 was achieved at a current density of 0.25 mA cm −2 . However, the laser fabrication parameters were selected experimentally and resulted in low performance of pseudocapacitors. Therefore, one of the main objectives of this study was to find the optimal laser fabrication parameters to achieve the highest specific areal capacitance. A large dataset was generated to find the relationship between the laser fabrication parameters and the electrochemical behavior performance (impedance and specific areal capacitance) of the fabricated electrodes by using an artificial neural network (ANN). We used an optimization algorithm (simulated annealing-SA) to overlook the trained ANN model as a black box and try to maximize the objective function, which in our case is a specific capacitance value, to find the most optimal laser fabrication parameters. Using SA, optimal laser fabrication parameters were found, which increased the specific areal capacitance to 0.9999 mF cm −2 at a current density of 0.25 mA cm −2 . The results demonstrated that the conducted study has the potential to introduce effective techniques for utilizing ULPING to produce nanoscale structures on TMs. These structures have the potential to be employed as electrodes in pseudocapacitors. Additionally, the research underscores the significance of employing data-driven approaches in electrode design procedures.

show abstract

“…The loss function used is MAE, one of the regression model evaluation metrics. MAE was chosen as the loss function because MAE is more resistant to data outliers [24]. The MAE principle considers the average difference in absolute value between the predicted result and the actual value.…”

Section: E Loss Functionmentioning

confidence: 99%

“…We used root mean square error (RMSE) and mean average error (MAE) to assess the model. Both assessment measures are used because RMSE performs well when examining model capabilities in certain circumstances and MAE performs better in others [24]. The smaller the MAE and RMSE values, the better the performance of the model.…”

Section: F Performance Evaluationmentioning

confidence: 99%

See 1 more Smart Citation

Transformer-based Deep Learning for COVID-19 Prediction Based on Climate Variables in Indonesia

Kurnianingsih¹,

Wirasatriya²,

Lazuardi³

et al. 2023

Int. J. Adv. Sci. Eng. Inf. Technol.

View full text Add to dashboard Cite

Recent research on the effect of climate variables on coronavirus (COVID-19) transmission has emerged. Climate change can potentially cause new viral outbreaks, illness, and death. This study contributes to COVID-19 disease prevention efforts. This study makes two contributions: (1) we investigated the impact of climate variables on the number of COVID-19 cases in 34 Indonesian provinces, and (2) we developed a transformer-based deep learning model for time series forecasting for the number of positive COVID-19 cases the following day based on climate variables in 34 Indonesian provinces. We obtained data from March 15, 2020, to July 22, 2021, on the number of positive COVID-19 cases and climate change variables (wind, temperature, humidity) in Indonesia. To examine the effect of climate change on the number of positive COVID-19 cases, we employed 15 scenarios for training. The experiment results of the proposed model show that the combination of wind speed and humidity has a weakly positive correlation with positive COVID-19 incidence; however, the temperature has a considerably negative association with positive COVID-19 incidences. Compared to the other testing scenarios, the transformer-based deep learning model produced the lowest MAE of 175.96 and the lowest RMSE of 375.81. This study demonstrates that the transformer model works well in several provinces, such as

show abstract

Root-mean-square error (RMSE) or mean absolute error (MAE): when to use them or not

Cited by 576 publications

References 24 publications

iupacGPT: IUPAC-based large-scale molecular pre-trained model for property prediction and molecule generation

iupacGPT: IUPAC-based large-scale molecular pre-trained model for property prediction and molecule generation

Optimizing the Ultrashort Laser Pulses for In Situ Nanostructure Generation Technique for High-Performance Supercapacitor Electrodes Using Artificial Neural Networks and Simulated Annealing Algorithms

Transformer-based Deep Learning for COVID-19 Prediction Based on Climate Variables in Indonesia

Contact Info

Product

Resources

About