Skip to content

Instantly share code, notes, and snippets.

Avatar

Ignasi Bartomeus ibartomeus

View GitHub Profile
@ibartomeus
ibartomeus / Parse terribly formated data
Last active Jun 2, 2020
Ideas on how to parse data stored as text with complex structure.
View Parse terribly formated data
#question: Can we create an heuristic to parse this type of data:
#Input example:
Halictus crenicornis
GALICIA: 1♀, Monte do Gozo, Santiago de Compostela (La Coruña), 350 m, 5.VIII.2016, 29TNH404481. – 1♀, Río Castro, Cerdedo (Pontevedra), 20.VII.1996. – 1♀, Oca (Pontevedra), 20.VII.1996.
ASTURIAS: 1♂, 1♀, Raitán, Carreño (Asturias), 130 m, 30TTP72, 17.VIII.2005. – 1♀, Poreño, Villaviciosa (Asturias), 43,426443º, -5,445950º, 13.V.2015, sobre flor de Centaurea nigra, C. Guardado leg. – 1♀, Poreño, Villaviciosa (Asturias), 43,426443º, -5,445950º, 27.V.2014, M. Miñarro leg. – 1♀, Muñiz (Asturias), 14.VII.2016, sobre flor de Taraxacum, D. Luna leg.
#Desired output (csv):
Species, CCAA, female, male, locality, province, elevation, date, UTM, latitude, longitude, notes
Halictus crenicornis, Galicia, 1, 0, "Monte do Gozo, Santiago de Compostela", La Coruña, 350, 5.VIII.2016, 29TNH404481, NA, NA, NA
@ibartomeus
ibartomeus / set.R
Last active Jan 2, 2018
Calculate how many possible sets there are in a set game
View set.R
#Calculate how many sets there are in a set game.
#each card has 4 characteristics with 3 factors each
color <- c("red", "green", "purple")
shape <- c("round", "diamond", "curly")
texture <- c("fill", "empty", "striped")
number <- c("one", "two", "three")
create_board <- function(n = 12){
View origin_demo.txt
#This is a quick demo performed for Sevilla R users group
#Elena asked about IUCN data. You can retrive this data using taxize: https://github.com/ropensci/taxize
#Elena also suggested vectorbase can be scrapped. Maybe for the next hackaton?
#package OriginR
library(originr)
#define species (don't worry about typos)
sp <- c("Apis mellifera", "carpobrotus edulis", "Lavandula stoechas",
@ibartomeus
ibartomeus / clean_species
Last active Jan 23, 2017
Cleaning species taxonomy using taxize. I want to correct synonyms and typo's and drop incomplete cases.
View clean_species
#I have >1000 bees to check its name, so I want to automatize taxize for
# fixing misspellings when possible
# updating synonims to accepted names
# keeping ONLY accepted species (fully resolved at species level)
# this uses taxize > 0.7.6.9157 If you are using older version (e.g. what its now on CRAN) see the history of this file.
library(taxize)
library(dplyr)
#example: good, synomin, typo, unexisting, genus only.
@ibartomeus
ibartomeus / ggplot
Created Nov 6, 2015
Código charlas SevillaR ggplot (Raúl Ortiz)
View ggplot
---
title: "Sevillarusers - ggplot2 intro"
author: "Ra?l Ortiz"
date: "Tuesday, October 27, 2015"
output: pdf_document
---
# Introducci?n al paquete gr?fico "ggplot2".
## Establezco el directorio de trabajo.
@ibartomeus
ibartomeus / PowerSevilla.R
Created Oct 16, 2015
Talk about power analtsis for the Sevilla R users meeting
View PowerSevilla.R
#SevillaR talk
#The problem:
time <- c(2000:2015)
abundance <- rnorm(16, 150, 50) #poison??
plot(abundance ~ time, t = "l")
#can I detect a trend?
View CIS
#Explore CIS data
#load data----
load("barometro_enero.RData")
head(barometro)
str(barometro)
head(nombres_etiquetas)
nombres_etiquetas
@ibartomeus
ibartomeus / random_slope_models
Created Jan 21, 2015
This shows how to get the random slopes and CI's for each level in a hierarchical model
View random_slope_models
#This shows how to get the random slopes and CI's for each level in a hierarchical model
#dataset used
head(iris)
#what we want to investigate
#Is there a general relationship? and how it differs by species
plot(iris$Sepal.Width ~ iris$Petal.Width, col = iris$Species, las =1)
#Our model with random slope and intercept
View pref
---
title: "Preferring a preference index"
author: "I. Bartomeus"
output: html_document
---
I've been reading about preference indexes lately, speciphically for characterizing pollinator preferences for plants, so here is what I learnt. Preference is defined as using an item (e.g. plant) more than expected given the item abundance.
First I like to use a quantitative framework (you can use ranks-based indices as in Williams et al 2011, which has nice propiertiest too). The simpliest quantitative index is the forage ratio:
@ibartomeus
ibartomeus / multifunc2
Created Nov 1, 2013
This approach to assess multi-functionality is based in the idea that sites providing best multiple functions will have not only a high mean value across function (approach 3 in Byrnes et al.) but also low variability in the function delivered across functions (i.e. Coef of var). It assumes you have read Byrnes et al paper (http://arxiv.org/abs/…
View multifunc2
# This approach to assess multifunctionality is based in the idea that sites providing
# best multiple functions will have not only a high mean value across function
# (approach 3 in Byrnes et al.) but also low variability in the function delivered
# across functions (i.e. Coef of var).
#I use Byrnes multifunc package to ilustrate it.
library(devtools)
install_github("multifunc", "jebyrnes")
library(multifunc)
library(ggplot2)
You can’t perform that action at this time.