MISA-XGModel 0.1.5

A Python library for predicting electron density using XGBoost

MISA_XGModel

A Python library for predicting electron density using XGBoost.

---

Overview

This library provides tools to predict electron density (Ne) using a machine-learning model based on XGBoost. It supports querying predictions for specific latitude, longitude, day of year (DOY), altitude, and solar local time (SLT), while leveraging pre-computed geophysical indices.

---

Features

• Predict electron density (Ne) for specific geospatial and temporal conditions.
• Clamp input values to the bounds of the training dataset for robustness.
• Supports querying geophysical indices from provided datasets.
• Modular design for integrating geophysical models in Python applications.

---

Installation

You can install this package via pip:

pip install MISA_XGModel

---

Usage

Quickstart

Here’s how you can use the library to predict electron density:

from MISA_XGModel import predict_ne, query_model

# Predict Ne with dataset lookup for geophysical indices
predicted_ne = predict_ne(
    lat=42.0, lon=-71.0, doy=99, alt=150.0, slt=12.0, year=2024
)
print(f"Predicted Ne: {predicted_ne:.2e}")

# Predict Ne with precomputed geophysical indices
predicted_ne = query_model(
    lat=42.0, lon=-71.0, doy=99, alt=150.0, slt=12.0,
    hp30=2, ap30=7, f107=209, kp=2.3, fism2=0.0007678
)
print(f"Predicted Ne: {predicted_ne:.2e}")

---

Inputs and Parameters

Clamping Input Values

Input parameters (lat, lon, doy, alt, slt) are clamped to the boundaries of the training data. These boundaries are defined as follows:

Parameter	Min Value	Max Value
lat	37.5	49.9
lon	-85.7	-76.1
alt	94.6 km	500 km
slt	0.0 hrs	24 hrs

---

predict_ne Function

Parameter	Type	Description
lat	float	Latitude of the input location (degrees). Clamped to [37.5, 49.9].
lon	float	Longitude of the input location (degrees). Clamped to [-85.7, -76.1].
doy	int	Day of year (1-365).
alt	float	Altitude in kilometers. Clamped to [94.6, 500].
slt	float	Solar local time (0-24 hours). Clamped to [0.0, 24].
year	int	Year to lookup geophysical indices in the dataset.
master_geo_ds	xarray.Dataset	Dataset containing geophysical indices (default: master_geo_ds).
model	XGBRegressor	Pre-trained XGBoost model for predictions (default: optimized_xgb).
scaler	StandardScaler	Scaler used for feature normalization (default: scaler_large).

This function also retrieves geophysical indices (hp30, ap30, f107, kp, fism2) from the provided dataset for the specified year and day of year (doy).

---

query_model Function

Parameter	Type	Description
lat	float	Latitude of the input location (degrees). Clamped to [37.5, 49.9].
lon	float	Longitude of the input location (degrees). Clamped to [-85.7, -76.1].
doy	int	Day of year (1-365).
alt	float	Altitude in kilometers. Clamped to [94.6, 500].
slt	float	Solar local time (0-24 hours). Clamped to [0.0, 24].
hp30	float	Precomputed geophysical index.
ap30	float	Precomputed geophysical index.
f107	float	Precomputed geophysical index.
kp	float	Precomputed geophysical index.
fism2	float	Precomputed geophysical index.
model	XGBRegressor	Pre-trained XGBoost model for predictions (default: optimized_xgb).
scaler	StandardScaler	Scaler used for feature normalization (default: scaler_large).

predict_ne requires geophysical indices as input, allowing for greater control over predictions in scenarios where geophysical data is already available.

---

Requirements

• Python 3.7+
• Dependencies:
  • xgboost
  • scikit-learn
  • numpy
  • pandas
  • xarray
  • joblib
  • netcdf4

---

License

This project is licensed under the MIT License - see the LICENSE file for details.

---

Author

• Mateo Cardona Serrano (them)
• GitHub Profile
• Email

---

Acknowledgments

• Thank you to Sevag Derghazarian for his continuous support and consultation on this project.

---