Jimmy Li fitst github pypi package
Project description
Modeling ring current proton distribution using artificial neural network (multi-layer perceptron).
This model is based on RBSP/RBSPICE measurements from 2013-2018. The RBSPICE measures proton flux at 14 energy channels from 45 keV to 598 keV, and this model provides proton spin-averaged flux in those 14 energy channels.
The 'lws_training_display_v1.0.ipynb' demonstrates the training of the model, as well as the illustration of comparision between test dataset and predictions over long-term. The processed data are stored in 'rcpann_model_files.zip'. Two additional subprograms are needed: 'time_double.string.py' which converts time formats and 'test_correlation.py' which plot comparison between test data and predictions.
To use this model, the users basically use 'rcpann_model_v1.0.ipynb', which load the model stored in 'rcpann_model_files.zip', and output the results (the log10 of flux in units of keV^-1c m^-2 s^-1). We also shows 4 examples of how to use this model. To simulate proton flux in a specific event, the uses also need the geomagnetic data stored in 'geomag_2012_2018.csv'.
The data for the training and modeling are available at https://doi.org/10.5281/zenodo.7651736.
Please contact jinxing.li.87@gmail.com for support.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for hellopkgqwe-0.0.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | b91cc99bf61f054a093b0c1732557553d67c97b312cce916ab005ed9b4b0e430 |
|
MD5 | 0166740cd664e0352e2f22ca8cc58f82 |
|
BLAKE2b-256 | a999e265e0e9f008e54efec8d60999098b22ada5beb7af0ce41cc9fadf3ce6d7 |