If you want to directly use our seismic features (either Type A or Type B) for your data,
you’ll need to prepare your seismic data as follows.
Please download the 0seismic_feature.zip and 3trained_model.zip from
Supporting material for “Enhancing debris flow warning via machine learning feature reduction and model selection”
Please
Unzip 0seismic_feature.zip and place the European folder into:
data/seismic_feature
Open config_I-O.yaml and relace the path “seismic_feature_dir”
# seismic data-60s source in the GFZ-GLIC server
"glic_sac_dir": "/path/to/your/3Diversity-of-Debris-Flow-Footprints/data/seismic_feature"
Unzip 3trained_model.zip and place the LSTM, Random_Forest, XGBoost folder into:
trained_model/v1-model
Open and run the inference tutorial notebook to get started with the model and explore its functionality.
Open config_I-O.yaml and replace the value for “glic_sac_dir”:
# seismic data source on the GFZ-GLIC server
"glic_sac_dir": "/storage/vast-gfz-hpc-01/project/seismic_data_qi/seismic"
Open config_I-O.yaml and relace the path “seismic_feature_dir”
# seismic data-60s source on the GFZ-GLIC server
"glic_sac_dir": "/storage/vast-gfz-hpc-01/home/qizhou/3paper/0seismic_feature""
glic_sac_dir/
└── European/ # continent
└── Illgraben/ # catchment
└── meta_data/ # sensor response
└── 9S_2017_2020.xml # seismic_network_year1_year2.xml
└── ...
└── 2020/ # year of the data
└── ILL12/ # station
└── EHZ/ # component
├── 9S.ILL12.EHZ.2020.001.mseed # seismic_network.station.component.year.julday.data_format
├── ...
└── 9S.ILL12.EHZ.2020.243.mseed
# Europen
Illgraben-9J:
path_mapping: "European/Illgraben"
response_type: "xml"
sensor_type: "do-not-need"
normalization_factor: "do-not-need"
sbatch calculate_features/sbatch/9S/submitStep1_2020.sh # remember change the parameter1, 2, 3
sbatch calculate_features/sbatch/9S/submitStep2_2020.sh # remember change the parameter1, 2, 3
sbatch calculate_features/sbatch/9S/submitStep3_2020.sh # remember change the parameter1, 2, 3
If you want to write your own code to calculate seismic features
calBL_feature in Type_A_features.py
calculate_all_attributes in Type_B_features.py