main.py

from __init__ import ROOT_DIR
import sys
sys.path.append(ROOT_DIR)

from list import LonList,LatList,GeoList
from ArgoData import Core
from TransGeo import TransitionGeo
import argparse
import argparse
import folium
import folium.plugins
import pandas as pd
import os
import geopy
import numpy as np
from TransMat import TransMat
from folium.plugins import HeatMap

class ExcelFloat(Core):
	@classmethod
	def from_excel(cls,filename):
		trans_mat = TransMat.load(os.path.join(ROOT_DIR,'90-2-2.npy'),GeoClass=TransitionGeo)
		file_path = os.path.join(ROOT_DIR,'Data/'+filename+'.csv')
		df = pd.read_csv(file_path,usecols=['Latitude','Longitude'])
		float_lats = LatList(df.Latitude.tolist())
		lat_bins = trans_mat.trans_geo.get_lat_bins()
		adjusted_lats = LatList(lat_bins.find_nearest(x) for x in float_lats)
		float_lons = LonList(df.Longitude.tolist())
		lon_bins = trans_mat.trans_geo.get_lon_bins()
		adjusted_lons = LonList(lon_bins.find_nearest(x) for x in float_lons)
		float_idxs = [trans_mat.trans_geo.total_list.index(geopy.Point(y,x)) for y,x in zip(adjusted_lats,adjusted_lons)]
		holder_array = np.zeros([len(trans_mat.trans_geo.total_list),1])
		for idx in float_idxs:
			holder_array[idx]+=1
		return (cls(holder_array,trans_geo=trans_mat.trans_geo),float_lats,float_lons)

parser = argparse.ArgumentParser()
parser.add_argument("filename", help="filename of csv file with argo locations",
                    type=str)
parser.add_argument("timestep", help="timestep of prediction",
                    type=float)
args = parser.parse_args()
assert args.timestep>45, "timestep must be greater than 45 days"
timestep = round(args.timestep/90)
if args.timestep-timestep*90>45:
	timestep += 1
float_array,float_lats,float_lons = ExcelFloat.from_excel(args.filename)
trans_mat = TransMat.load(os.path.join(ROOT_DIR,'90-2-2.npy'),GeoClass=TransitionGeo)
trans_calc = trans_mat.multiply(timestep-1)

output = trans_calc.todense().dot(float_array.todense())
output = [x[0] for x in output.tolist()]
lats,lons = zip(*trans_mat.trans_geo.total_list.tuple_total_list())
df = pd.DataFrame({'Latitude':lats,'Longitude':lons,'Probability':output})
df = df[df.Probability!=0]
df = df.reset_index(drop=True)

url = 'https://server.arcgisonline.com/arcgis/rest/services/Ocean/World_Ocean_Base/MapServer/tile/{z}/{y}/{x}'
map=folium.Map(location=[0,0],zoom_start=2)
folium.TileLayer(tiles=url, attr='World_Ocean_Base').add_to(map)
for x,y in zip(float_lons,float_lats):
	folium.Marker([y,x],
	              popup='Float Located at %s , %s'%(y,x),
	              icon=folium.Icon(color='green')
	             ).add_to(map)	
HeatMap(df, 
        min_opacity=0.4,
        blur = 18
               ).add_to(folium.FeatureGroup(name='Heat Map').add_to(map))
folium.LayerControl().add_to(map)


df.to_csv(os.path.join(ROOT_DIR,'Output/'+args.filename+'_tm_output.csv'))
map.save(outfile=os.path.join(ROOT_DIR,'./Output/map.html'))
os.system('open '+os.path.join(ROOT_DIR,'./Output/map.html'))