old.txt

 ## Sources of data

 ### [KP] (https://kp.gfz-potsdam.de/en/about-kp)
 ### proton files 1994..2016 
 - [MIT Space Plasma Group](https://web.mit.edu/space/www/wind/wind_data.html)
 - ftp://space.mit.edu/pub/plasma/wind/kp_files/
   This data set has about 9857 Files (one per day each Mission day). Each file has 360 records of 15 fields each. it is 1.3 GB.
#### Fields

  - year  
  - day (January 1 = 1) 
  - fraction_of_day  
  - speed (km/s)  
  - E/W_angle (deg) see below 
  - N/S_angle (deg) see below 
  - proton_density (#/cc) 
  - most probable proton_thermal_speed (km/s)  {=sqrt[2kT/Mp] }  
  - alpha_% = 100*(alpha number dens)/(proton number dens) 
       {fill value (-1.e+31) if not available} 
       {This quantity will be added in future analyses.}
  - Vx (km/s)  (velocity component in GSE)  
  - Vy (km/s)  (velocity component in GSE )
  - Vz (km/s)  (velocity component in GSE)
  - XSE (Re) s/c position
  - YSE (Re) s/c position
  - ZSE (Re) s/c position

### Solar images
[**So**lar & **H**eliospheric **O**bservatory mission -- SOHO --](https://www.nasa.gov/mission_pages/soho/index.html) 

[THE ACE SCIENCE CENTER](https://izw1.caltech.edu/ACE/ASC/)
[SPACE WEATHER PREDICTION CENTER NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION](https://www.swpc.noaa.gov/products/lasco-coronagraph)


# Planning
## Goal
Elaborate a friendly usage predicition machine 
### EPICS
0.- Define full stack
1.- Identify data series
2.- Automatize training data and new data upload
3.- Data normalization
4.- MOdel training
5.- model validation
6.- BUild Dashboard

[data cleaning](https://github.com/luismiguelcasadodiaz/Helio_Sentinel_Crew/blob/main/Data_Cleaning.md)