reachsumit

reachsumit

reachsumit

reachsumit

reachsumit

reachsumit

reachsumit

Area of Interest, Satellite, Location, Panchromatic Resolution, Area of Raster (Sq Km), Building Labels (Polygons), Road Labels(LineString), Training Data, Test Data
AOI_1, WorldView-2, Rio de Janeiro, 50 cm, 2544, 4082529, N/A, 3-band Mosiac = 2.3 GB | 8-band Mosiac = 6.5 GB | Vectors = 0.18 GB | Point of interest = 31 GB, N/A
AOI_2, WorldView-3, Vegas, 30 cm, 216, 151367, 3685 km, Roads: 25 GB of 400mx400m tiles | Buildings: 23 GB of 200mx200m tiles, Roads: 8.1 GB of 400mx400m tiles | Buildings: 7.9 GB of 200mx200m tiles
AOI_3, WorldView-3, Paris, 30 cm, 1030, 23816, 425 km, Roads: 5.6 GB of 400mx400m tiles | Buildings: 5.3 GB of 200mx200m tiles, Roads: 1.9 GB of 400mx400m tiles | Buildings: 1.8 GB of 200mx200m tiles
AOI_4, WorldView-3, Shanghai, 30 cm, 1000, 92015, 3537 km, Roads: 25 GB of 400mx400m tiles | Buildings: 23.4 GB of 200mx200m tiles, Roads: 8.1 GB of 400mx400m tiles | Buildings: 7.7 GB of 200mx200m tiles
AOI_5, WorldView-3, Khartoum, 30 cm, 765, 710960, 1030 km, Roads: 25 GB of 400mx400m tiles 

reachsumit

from scipy.io import loadmat

# Load label information
matrix = loadmat("../cars_annos.mat")

# create as many directories as the number of classes in dataset (196)
import os
for idx in range(matrix['class_names'].shape[1]):
    os.makedirs("./"+str(idx+1))

reachsumit

import pandas as pd

input_df = pd.read_csv(r"merged.csv")
state = ""
weights = []
x_numerator = []
y_numerator = []
results = pd.DataFrame(data=None,columns=['State','Longitude','Latitude'])

for index,x in input_df.iterrows():

reachsumit

from bs4 import BeautifulSoup as bs
import requests

# Base url for scraping
res=requests.get('http://www.citypopulation.de/India.html')
# delicious soup
soup= bs(res.text,'lxml')
# path to the required links
allLinks = soup.select('ul li a')
# keep only the links with full district data