Tom Hengl - Step-by-step tutorial to optimization of geocomputing (tiling & parallelization) with R'

preview_player
Показать описание
Lecturer: Tomislav Hengl (OpenGeoHub foundation)

Summary: Large datasets e.g. large rasters or vectors can not be easily processed in R, mainly because of the RAM and storage limitations. In addition, within a High Performance Computing environment we are interested in running operations efficiently in parallel (e.g. using snowfall, future or similar packages), so that the total computing time can be minimized. Within the landmap package several functions are now available to tile large rasters and then run processing by applying custom based function per tile in parallel. This is at the order of magnitude more complex than running functions on the complete object, hence in a step-by-step tutorial I will try to explain how to make your own custom functions and combine R, GDAL, SAGA GIS and QGIS to run processing and eventually also Cloud-Optimized GeoTIFFs to share large datasets.

References:
Conrad, O., Bechtel, B., Bock, M., Dietrich, H., Fischer, E., Gerlitz, L., … Böhner, J. (2015). System for automated geoscientific analyses (saga) v. 2.1.4. Geoscientific Model Development, 8(7), 1991–2007. doi:10.5194/gmd-8-1991-2015
Mitchell, T., & GDAL Developers. (2014). Geospatial Power Tools: GDAL Raster & Vector Commands. Locate Press.

How to cite this video:
  1. Tomislav Hengl (OpenGeoHub Foundation)
  2. OpenGeoHub
  3. R spatial
  4. Open GeoData
  5. Geospatial Data
Рекомендации по теме
Tom Hengl - 1:39:37

Tom Hengl - Step-by-step tutorial to optimization of geocomputing (tiling & parallelization) wit...

Introduction to Ensemble 0:46:26

Introduction to Ensemble Machine Learning for Predictive Soil Mapping (landmap package) part III

Introduction to Ensemble 1:26:06

Introduction to Ensemble Machine Learning for Predictive Soil Mapping (landmap package) part I

Introduction to Ensemble 1:27:04

Introduction to Ensemble Machine Learning for Predictive Soil Mapping (landmap package) part II

Tom Hengl: 'Machine 0:51:39

Tom Hengl: 'Machine Learning as a generic framework for spatial prediction'

Tomislav Hengl: 'Automated 1:57:07

Tomislav Hengl: 'Automated predictive mapping using Ensemble Machine Learning'

Tomislav Hengl - 1:57:05

Tomislav Hengl - Automated predictive mapping using Ensemble Machine Learning

Tomislav Hengl: 'Computing 1:19:50

Tomislav Hengl: 'Computing with large rasters in R: tiling, parallelization, optimization'

Tomislav Hengl: 'Mapping 1:25:30

Tomislav Hengl: 'Mapping species distribution using Machine Learning'

Hengl, T., Bonannella, 1:31:50

Hengl, T., Bonannella, C.: 'Spatiotemporal Ensemble ML in R' training session

OEMC 19 July 0:23:09

OEMC 19 July 2022 1 Tom Hengl wQ&A

Ichsani Wheeler and 0:03:06

Ichsani Wheeler and Tom Hengl – Open source, satellites and data to know what happens on our planet...

Open Data Science 2:40:36

Open Data Science Europe workshop 2022 Day 3

GEOSTAT 2014 Bergen 1:26:24

GEOSTAT 2014 Bergen Day 4: plotKML tutorial (screen 1)

Pedometrics Brazil 2021: 0:56:30

Pedometrics Brazil 2021: Computing with Cloud-Optimized GeoTIFFs in R

Hengl, T. 'Introduction 1:24:36

Hengl, T. 'Introduction to spatial and spatiotemporal data in R'

OpenGeoHub Summer School 2:01:36

OpenGeoHub Summer School 2021 Day 1 - September 1 - block 1

Christian Knoth - 1:49:24

Christian Knoth - Introduction to Deep Learning in R for analysis of UAV-based remote sensing data

Tom Hengl: OpenLandMap 1:25:40

Tom Hengl: OpenLandMap - Global ARCO environmental layers

OpenGeoHub Summer School 1:02:24

OpenGeoHub Summer School 2021 Day 1 - September 1 - Block 2

LandGIS in action 0:05:55

LandGIS in action --- web-mapping app to explore and download environmental data

OpenGeoHub Summer School 1:38:37

OpenGeoHub Summer School 2022 - Room 2 - 1:30 to 3 pm

GEOSTAT 2014 Bergen 0:51:53

GEOSTAT 2014 Bergen Day 1: Introduction to GEOSTAT

Martijn Tennekes: Plenary 0:17:55

Martijn Tennekes: Plenary - 03.09.2019