M3nin0

# Adapted from: https://github.com/albhasan/deforestation-sentinel2/blob/master/06_validation/validate_results.R
# validation_dataset_bands.csv is avaliable online on: https://doi.pangaea.de/10.1594/PANGAEA.921387?format=html#download
set.seed(666)

validation_dataset_bands <- readr::read_csv("validation_dataset_bands.csv")

deforestation <- dplyr::filter(validation_dataset_bands, label == "Deforestation")
non_forest    <- dplyr::filter(validation_dataset_bands, label == "Non-Forest")
forest        <- dplyr::filter(validation_dataset_bands, label == "Forest")

M3nin0

default:

    # estimated relative growth size of R memory relative to block size
    R_memory_bloat           : 4
    R_processing_bloat       : 5

    # minimum number of samples for multicore processing
    minimum_number_samples   : 1000

    # Data cube types known to sits

M3nin0

chooseCRANmirror(graphics=FALSE, ind=1)

# Install devtools, rmarkdown, knitr, testthat and Rcpp if not already available
install.packages(c("devtools", "rmarkdown", "Rcpp", "knitr", "testthat", "remotes"))

# Please install the Suggested packages that are used by sits
install.packages(c("DBI","dendextend", "dtwclust","dtwSat", "e1071", "flexclust",
                  "imager", "imputeTS", "kohonen", "lwgeom", "MASS", "methods",
                  "mgcv", "nnet", "proto", "proxy", "ptw", "ranger", "RCurl",
                  "RSQLite", "signal", "xgboost", "zoo", "randomForest", "desc"))

M3nin0

version: 0.6.0
inputs:
    directories:
        - R
    options:
        CACHE: on # Cache the steps results
workflow:
    type: serial
    specification:
        steps:

M3nin0

# STAC.py
bdc_stac_service = stac.STAC('http://brazildatacube.dpi.inpe.br/stac/')

# Select a collection and get items
collection = bdc_stac_service.collection('S2_10_16D_STK-1')
items = collection.get_items(filter={'bbox':'-61, 2.8, -60, 1.8', 'datetime':'2018-08-01/2019-07-31'})

# Generate a "1D" (One variable) datacube
## mutate_1d_cube from github.com/M3nin0/xarray-stac/
cube1d = mutate_1d_cube(items['features'], 'NDVI')

M3nin0

# Autor: Felipe Menino Carlos
# Data: 11/09/2020
'
Calcula o SpeedUp
'
speedUp <- function(serialTime, parallelTime)
{
  return(serialTime / parallelTime)
}

M3nin0

import stac

bdc_stac_service = stac.STAC('http://brazildatacube.dpi.inpe.br/bdc-stac/0.8.0/')
collection = bdc_stac_service.search({'collections': ['LC8SR'], "intersects": {"type": "Point", "coordinates": [-45, -13] }, "limit": 10})

M3nin0

import numpy as np
import pandas as pd
from sklearn import linear_model
import matplotlib.pyplot as plt

# Carregando os dados
data_owd = pd.read_csv('https://covid.ourworldindata.org/data/owid-covid-data.csv')

# Filtrando os dados (Canadá)
can = data_owd[data_owd.iso_code == 'CAN']

M3nin0

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%       Arbitrary Spectral Slope Noise Generation      %
%           with OCTAVE Implementation                 %
%                                                      %
% From Little, 1992. Adapted from R. Rosa (08/08/14)   %
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

function x = powernoise(beta, N, varargin)
% Generate samples of power law noise. The power spectrum
% of the signal scales as f^(-beta).

M3nin0

#include <iostream>
// Idea from: https://stackoverflow.com/questions/5171739/tostring-override-in-c

class Teste
{
        private:
        int a;

        public:
        Teste(int a)