GeoComp & ML 2024 course

Geocomputation and Machine Learning for environmental applications

(Spatial Ecology is supporting the RAM4Africa project)

Online teaching: April and June 2024, 10 weeks

  • Start lecturing: 2 April, every Tuesday, Thursday at 3 pm - 5:45 pm UTC (CEST 5 pm, EDT 11 am, PDT 8 am)

  • Catch-up session: Tuesday 30 April: 3 pm - 5:45 pm UTC (CEST 5 pm, EDT 11 am, PDT 8 am)

  • Catch-up session: Thursday 2 May: 3 pm - 5:45 pm UTC (CEST 5 pm, EDT 11 am, PDT 8 am)

  • Resume lecturing: 7 May, every Tuesday, Thursday at 11 am - 1:45 pm UTC (CEST 1 pm, EDT 7 am, PDT 4 am)

  • GRASS week: 21-22-23 May, Tuesday, Wednesday, Thursday at 11 am - 1:45 pm UTC (CEST 1 pm, EDT 7 am, PDT 4 am)

  • End lecturing: 6 June, at 11 am – 1:45 pm UTC (CEST 1 pm, EDT 7 am, PDT 4 am)

In-person week: June 10-14, coding hackathon in Matera, Italy

In-person week: June 17-21, extended study group in Matera, Italy

The in-person Matera session can be considered a sort of hackathon where participants will work on their code from 9am to 5pm and on the last days will present their final project and code. This week aims to achieve group collaboration and troubleshooting under the direct supervision of the trainers. During this session, there will also be brief talks from the trainers and from invited speakers.

Course objectives

In this course, students will be introduced to an array of powerful open-source geocomputation tools and machine learning methodologies under Linux environment. Students who have never been exposed to programming under Linux are expected to reach the stage where they feel confident in using very advanced open source data processing routines. Students with a precedent programming background will find the course beneficial in enhancing their programming skills for better modelling and coding proficiency. Our dual teaching aim is to equip attendees with powerful tools as well as rendering their abilities of continuing independent development afterwards. The acquired skills will be beneficial, not only for GIS related application, but also for general data processing and applied statistical computing in a number of fields. These essentially lay the foundation for career development as a data scientist in the geographic domain.

All the class will be recorded. The video link will be posted in the syllabus below

Class roster

Syllabus overview

Subject n. lectures
LINUX: 1
BASH/AWK: 2
GDAL/PKTOOLS: 4
GEO-PYTHON: 3
GRASS: 3
ML: 6

Detailed syllabus

Below we are going to list day by day the final syllabus with the relatives links to the materials and recorded video.

Lecture 1: Tuesday, 2nd of April, 2024.

Getting started: knowing each other and course introduction.

This session introduces the objective of the course and a round table among the students and the teachers.

Recorded lecture can be found here1 and here2

Time below in CEST

Catch-up session 0: Wednesday, 3rd of April, 2024

This session is dedicate to the students that encounter problems for the OSGeoLive 16.0 Virtual Machine installation. Be sure that following OSGeoLive 16.0 Virtual Machine features are present:

  • The OSGeoLive 16.0 Virtual Machine internet connection is working fine - open a browser and search for something.

  • The OSGeoLive 16.0 Virtual Machine screen is wide and you can expand and switch back to your native OS without issues.

  • The shared folder has been correctly installed. Check if the /media/sf_LVM_shared is visible.

  • The /home/user/SE_data/ has been download and that the /media/sf_LVM_shared/my_SE_data is visible and you can see the sub-folder.

  • The OSGeoLive 16.0 Virtual Machine Keyboard Layout has been set in accordance to your keyboard, and that the letter that you type correspond in what you see.

If one or more of these features are not working properly show up in the Catch-up session.

Lecture 2: Thursday, 4th of April, 2024.

Jump start into LINUX Bash programming (Giuseppe Amatulli)

During this session we explore and practice the basics of BASH terminal command line. The acquired skills will be used in all following sections.

Recorded lecture can be found here.

Suggested assignments:

These assignments do not need to be delivered nonetheless we suggest doing it in order to get familiar with the bash command line topics. Below, some suggestions of possible operations:

  • Get fast in navigating up and down in the directory using only the command line.

  • Copy/move files from a directory to another directory.

  • Search and use a command for copy file from you pc to a Linux remote server.

  • Get fast on using keyboard shortcuts.

  • Perform a data download operation using wget or curl.

  • Unzip/untar files.

  • Get familiar with apt update && apt upgrade && apt install to maintain your OSGeoLIve Linux Virtual Machine and install new software.

Lecture 3: Tuesday, 9th of April, 2024.

Discovering the power of AWK programming language (Giuseppe Amatulli).

Recorded lecture can be found here.

This session is fundamental for data filtering and preparation, bulk data download, text files manipulation, descriptive statistics and basic mathematical operation on large files. Students will access, query, understand and clean up data, perform data filtering using bash command line. We use AWK which is an extremely versatile and powerful programming language for working on text files, performing data extraction and reporting or to squeeze data before importing them into R/python or other software types.

  • Welcome to AWK world. Why to use AWK command line (Lecture).

  • The basic commands, command syntax (Hands on tutorial).

    • Built in variables

    • Import variables

    • String functions

    • Numerical functions

    • Query functions

    • Manipulate large files before importing in R

    • A study case: Manipulate thousand of text file with BASH and AWK

    • This session summarizes the use of BASH and AWK with a practical example of text file manipulation.

  • Explain GSIM text files (Listen to the recorded video lecture and proceed with self-learning mode.)

    • Data exploration (Hands on tutorial).

    • Count number of observations

    • Count how many observations per date

    • Monthly MEAN distribution

Suggested assignments:

Please send the jupyter file or sh file (name as name_surname.ipynb or name_surname.sh ) as e-mail attachment to g.amatulli@spatial-ecology.net.

Using the bash (and/or awk) language manipulate the GSIM/US*.mon files in order to create txt files, one for each month-year, that includes the station ID, latitude, longitude and the MEAN value.

The final output will be n text files (form FirstDateOfTheSeries to LastDateOfTheSeries) with the following structure:

cat 2002-01.txt

Gsim.no latitude longitude MEAN
US_0001971 33.79427255 -84.4743747 0.916785714285714
US_0001977 33.65666667 -84.6736111 74.5558064516129

In case you wanna manipulate your own txt (or csv) files, fell free to do it by implementing everything in bash (and/or awk). You can perform:

  • data summary

  • table re-organization

  • data filtering

cd /media/sf_LVM_shared/GSIM/zip/GSIM_indices/TIMESERIES/monthly
cp /home/user/SE_data/exercise/manipulate_GSIM_assignment_solutions.sh /media/sf_LVM_shared/GSIM/zip/GSIM_indices/TIMESERIES/monthly

##### create output dir 
mkdir -p output1   output2 output3

#########Solution 1 #########*

rm -f output1/*.txt

# Creates a txt file with the list of all uniq dates

awk -F , '{ if(NF>5) { if ($1 > 0) { print $1 }}}'  US_00090*.mon  | sort | uniq > output1/dates.txt

# Creates a txt file with the list of all uniq station ID, longitude, latitude. 



paste -d " " <(grep gsim.no   US_00090*.mon | awk '{print $4}') \
         <(grep longitude US_00090*.mon | awk '{print $4}') \
         <(grep latitude  US_00090*.mon | awk '{print $4}') | sort -k 1,1 > output1/ID_x_y.txt

## loop trough the dates.txt 

for DATE in $(cat output1/dates.txt ) ; do 
echo processing $DATE

grep ^$DATE US_00090*.mon  | awk '{ gsub(":"," "); gsub(","," ");if($3!="NA"){print substr($1,1,10),$3}}' | sort -k 1,1 > output1/${DATE}_ID_mean.txt 

join -1 1 -2 1 output1/ID_x_y.txt output1/${DATE}_ID_mean.txt  > output1/${DATE}_ID_x_y_mean.txt

rm output1/${DATE}_ID_mean.txt

done

#########Solution 2 #########

rm -f output2/*.txt 
#loop over all US files
n=0
for i in US_00090*.mon ; do
#pull out information of interest
gsim=$(awk 'NR==11 {print $4}' $i)
lat=$(awk 'NR==15 {print $4}' $i)
long=$(awk 'NR==16 {print $4}' $i)

#make new directory with name of gsim ID to store txt files in
mkdir $output$gsim

#awk uses these and spits out txt files named by the date
n=$(expr $n + 1)
echo $gsim $n

awk -v gsim=$gsim -v lat=$lat -v long=$long -v output=$output \
' NR>22 {gsub(","," "); if($2!="NA"){print gsim,long,lat,$2 >> "output2/"$1"_ID_x_y_mean.txt"}}' $i

rm -r $gsim
done

#########Solution 3 #########

rm -f output3/*.txt

awk '{gsub(",", "");
 gsub("\t", " ");
 if($2=="gsim.no")   {station=$4};
 if($2=="latitude")  {lat=$4};
 if($2=="longitude") {lon=$4};
 if($1 ~ /[0-9][0-9][0-9][0-9]-[0-9][0-9]-[0-9][0-9]/ && $2!="NA" )
 { print station, lat , lon , $2 > "output3/"substr($1,1,7)"_ID_x_y_mean.txt" }
 }' US_00090*.mon

#########Solution 4 #########

From Issam Boukhris; it allows the closing of the file and can be run in the VM

sudo bash -c 'awk -F "," '\''{
    if ($1 ~ "gsim.no") split($1, a, ":");
    station = a[2];
    if ($1 ~ "longitude") split($1, b, ":");
    long = b[2];
    if ($1 ~ "latitude") split($1, c, ":");
    lat = c[2];
    if (NF>5 && $1 != "NA" && $2 != "NA") {
        print station, long, lat, $2 >> "output1/"substr($1,1,7)".txt";
        close("output1/"substr($1,1,7)".txt"); print echo substr($1,1,7)
    }
}'\'' US_*'

Lecture 4: Thursday, 11th of April, 2024.

Manipulate geographical data with GDAL/OGR (Giuseppe Amatulli).

Recorded lecture can be found here.

This section introduces data manipulation for geospatial data processing on the command line using GDAL & OGR libraries.

Suggested assignments:

Get familiar with gdalinfo & gdal_edit.py using your data or the tif file stored in the SE_data.

Watch this lecture to get more insight on the different data type
* Numerical System Decimal vs Binary. Recorded lecture: 1:28:45 - 1:48:00

These assignments do not need to be delivered, nonetheless we suggest doing it in order to get familiar with gdal_calc.py and with the issues of selecting the correct datatype and gdal_calc formula.

Lecture 5: Tuesday, 16th of April, 2024.

Manipulate geographical data with GDAL/OGR (Giuseppe Amatulli).

Recorded lecture can be found here.

This section introduces data manipulation for geospatial data processing on the command line using GDAL & OGR libraries.

Compulsory assignments:

  • This assignment focus in attending a video recording lecture aiming to push the precessing in the multi-core environment.

  • Multicore operation within gdal and pktools

    • Transform a simple “for loop” in multicore “for loop”

Suggested assignments:

  • These assignments do not need to be delivered now, nonetheless do it in order to get familiar with the gdal/ogr command line topics. In two weeks, when all the geographic tools will be covered, you will need to handling a scripting procedure that prepare the geographic dataset for your final project, so better to start now :-).

Below, some suggestions of possible operations:

  • Start to prepare the geographic dataset for your final project.

    • Using wget and/or curl download geographic dataset from an official repository

    • Assess and re-align geographic extent and pixel resolution (gdal_edit.py)

    • Using gdal command performs: cropping, merging (vrt+gdal_translate), re-projecting.

Lecture 6: Thursday, 18st of April, 2024.

Manipulate geographical data with PKTOOLS (Giuseppe Amatulli)

Recorded lecture can be found here.

This section introduces data manipulation for geospatial data processing on the command line using PKTOOLS.

Compulsory assignments:

Before the 23rd of April, 2024, watch the “Manipulate Landsat data with GDAL and PKTOOLS for a gap filling example” lecture (all links in Lecture 7). On the 23rd of April, 2024 we will do a QA and a review of gdal and pkstools commands. Feel free to present your scripting procedure for receiving feedback and suggestions.

Suggested assignments:

  • Assess and re-align geographic extent and pixel resolution (gdal_edit.py)

  • Using gdal command performs: cropping, merging (vrt+gdal_translate), re-projecting.

  • Using pktools command performs: masking, filtering, histogram, re-classification, zonal statistic.

Lecture 7: Tuesday, 23rd of April, 2024.

Manipulate Landsat data with GDAL and PKTOOLS for a gap filling example (Giuseppe Amatulli).

Video Recording of the lecture

Recorded of the QA can be found here.

This section introduces a real example of satellite data GLAD ARD processing using GDAL/OGR and PKTOOLS in a Bash environment.

Lecture 8: Thursday, 25th of April, 2024.

Case study presentation: Estimation of tree height (Giuseppe Amatulli)

Recorded lecture can be found here.

Compulsory assignments:

Catch-up session 1: Tuesday, 30th of April, 2024

Students will meet online with the teachers to review past material and solving additional doubts.

Recorded of the QA can be found here. Password has been sent by e-mail on March 28 (it will be the same pass-code for all recordings).

Catch-up session 2: Thursday 2nd of May, 2024.

Students will meet online with the teachers to review past material and solving additional doubts.

Recorded of the QA can be found here. Password has been sent by e-mail on March 28 (it will be the same pass-code for all recordings).

Lecture 9: Tuesday, 7th of May, 2024.

Geospatial python (Francesco Lovergine)

Recorded lecture can be found here.

Lecture 10: 9th of May, 2024.

Geospatial python (Francesco Lovergine)

Recorded lecture can be found here.

Lecture 11: Tuesday, 14th of May, 2024.

Machine Learning Introduction, Support Vector Machine & Random Forest (Antonio Fonseca & Giuseppe Amatulli)

In this class we will set the main fundamentals to implement machine learning in geo science. We will use the data-set and explanation described in Estimation of tree height using GEDI dataset - Data explore.

Recorded lecture can be found here.

  • Intro to machine learning (F.)

    • Defining learning

    • Supervised vs Unsupervised learning

    • The framework of learning algorithms

  • Example of Supervised learning using scikit-learn in python.

Lecture 12: Thursday, 16th of May, 2024.

Machine Learning - Perceptron (Antonio Fonseca)

Recorded lecture can be found here.

Lecture 13: Tuesday, 21st of May, 2024.

Introduction to GRASS (1) (Giuseppe Amatulli)

  • GRASS architecture

    • GRASS GUI

    • GRASS command structure

    • GRASS working environment

    • GRASS Location/Mapset

    • GRASS scripting procedure in BASH

Lecture 14: Wednesday, 22nd of May, 2024.

Geocomputing with grass GRASS (2) (Giuseppe Amatulli)

  • GRASS computation

    • Importing and exporting raster/vector

    • Buffer operation

    • Map calc in grass

    • Raster statistic

Lecture 15: Thursday, 23rd of May, 2024.

Hydrology analysis using GRASS (3) (Giuseppe Amatulli)

Lecture 16: Tuesday, 28st of May, 2024.

Machine Learning - Perceptron (Antonio Fonseca)

Recorded lecture can be found here.

Lecture 17: Thursday, 30th of May, 2024.

Machine Learning - Architecture of Neural Networks (Antonio Fonseca)

Recorded lecture can be found here.

Lecture 18: Tuesday, 4th of June, 2024.

Machine Learning - Capacity, Overfitting, Underfitting - Convolutional Neural Networks (Antonio Fonseca)

Recorded lecture can be found here.

  • Feedforward Neural Networks

    • Quick recap

    • Extra regularization techniques

    • Capacity, Overfitting and Underfitting

    • Debugging tips

    • Family of optimizers

    • Tutorial: more features and different optimizers

  • Convolutional Neural Networks

    • Kernels, padding, pooling

    • Classification tasks

    • Tutorial: data batching, classification of satellite images

    • Kernels, padding, pooling

    • Classification tasks

    • Tutorial: data batching, classification of satellite images

Lecture 19: Thursday, 6th of June, 2024.

Machine Learning - Papers discussion (Antonio Fonseca)

Recorded lecture can be found here.

In-person week: June 10-14, coding hackathon in Matera, Italy