Relationships between fault friction, slip time, and physical parameters explored by experiment-based friction model: A Machine Learning Approach Using Recurrent Neural Networks (RNNs)

1. 'Raw Data (HVR1087)': This dataset contains raw experimental data from the HVR1087 experiment. It includes parameters such as slip velocity, accumulated slip, axial shortening, normal stress, shear stress, temperature, and friction. The units for each parameter are provided within the file. The file is formatted as a CSV (Comma eparated Values) file. 2. 'Analyzed Data (HVR1087)': This file consists of data that has been processed and used for machine learning analysis in the current study. The parameters contained in this file, including Average slip velocity, Average axial shortening, Average normal stress, Average shear stress, Average friction, Average temperature, Rate of temperature, Rate of axial shortening, and Power density, were calculated based on the 'Raw data (HVR1087)'. The units for each parameter are provided within the file. This data is also formatted as a CSV file. 3. 'Constructing Model': This file contains the script or code used to carry out machine learning analysis in this study. It includes procedures such as data preprocessing, data splitting, construction of the training model, plotting of validation and test data, and tracing of gradient importance. Additionally, it provides information regarding the hyperparameters used in the study ('#ref'). The file format for this data is .ipynb, and it is compatible with Jupyter Notebooks.