🌱 Google Earth Engine 101

33rd National Geoscience Conference 2021 (NGC 2021)

Tool: ArcGIS Pro, ArcGIS Pro Deep Learning extension, Python, Jupyter Notebook Technique: Deep learning; semantic segmentation, cartography, remote sensing

The presentation of abstract outlining the implementation of deep learning in land cover classification across the Borneo island. It uses the Sentinel-2 image data and the band combination that differentiates the bareland, tree cover as well as waterbodies and croplands whilst training the U-Net model using the referenced data collected.

Please find the abstract published here:

Warta Geologi, Vol. 47, No. 1, April 2021

The presentation slide can be accessed at the following link 👇🏻:

Yes peeps. I’ve been studying and on contrary to all my previous attempts to make beautiful notes, I say f it and just work with what helps me clear my head the fastest 🏃🏻‍♀️. I love writing notes, but I realize, to gather my thoughts properly, I need some sort of way to not waste paper just to arrange and rearrange my ideas or comprehension of things. 

What better way of doing that than using a mind map!

So you kiddos out there who are starting out with Python and just can’t wait to get into deep learning or machine learning, I’d say, hold your horses for a minute and have some preview of that pond you’re trying to jump into. And don’t be scared, cause we’re all friends here in the hell-hole of learning plateau. Will it get better? I believe so. I am positive I understand more of the principles of deep learning and the relevance of Python libraries associated with it. Yes...this is a Python bar, darling. 👩🏻‍💻

There’s no real shortcut if you ask me since we have different way of comprehending things; my pre-existing mold may have harder time grasping the things I am learning right now than you would. So don’t be afraid to doodle while you think. No amount of paper will be enough to help you understand things, so better start being sustainable by using some digital platforms and saving those papers to when you’re truly ready to pen out your understanding of things; not what you read. There’s a difference!

Check out the mind map of some essential Python libraries you can get started with before you start doing some deep learning. It’s worth reviewing all that prior, I promise. 

Have fun! 🙆🏻‍♀️

📑 International Climate Initiative (IKI) Land Use Plan: Green Initiative in the Heart of Borneo (HoB) Report

Tool: ArcGIS Pro 2.9.3 Technique: Overlay analysis, visualization via remote sensing technique

These maps are developed to aid or supplement the Natural Capital Valuation (NatCap) initiative. As cited by WWF:

An essential element of the Natural Capital Project is developing tools that help decision makers protect biodiversity and ecosystem services.

One of the site included in this initiative by WWF-Malaysia is the Heart of Borneo (HoB). Specifically for this exercise, the visualization of policy and land use eventually become the data input utilized in the tool InVest that generates the models and maps for the economic values of ecosystem services within the landscape of interest.

The generation of the data mainly includes superficial remote sensing to assess the status of the land use in the respective concessions using Sentinel-2 satellite image with specific band combination to identify tree cover, particularly mangrove forest.

zero to pandas

There are alot of Python courses out there that we can jump into and get started with. But to a certain extent in that attempt to learn the language, the process becomes unbearably long and frustratingly slow. We all know the feeling of wanting to run before we could learn how to walk; we really wanna get started with some subtantial project but we do not know enough to even call the data into the terminal for viewing.

Back in August, freeCodeCamp in collaboration with Jovian.ai, organized a very interesting 6-week MOOC called Data Analysis with Python: Zero to Pandas and as a self-proclaimed Python groupie, I pledged my allegiance!

If there are any expectation that I've managed to whizz myself through the course and obtained a certificate, nothing of that sort happened; I missed the deadline cause I was busy testing out every single code I found and work had my brain on overdrive. I can't...I just...can't. Even with the extension, I was short of 2 Pythonic answers required to earn the certificate. But don't mistake my blunders for the quality of the content this course has to offer; is worth every gratitude of its graduates!

Zero to Pandas MOOC is a course that spans over 6 weeks with one lecture webinar per week that compacts the basics of Python modules that are relevant in executing data analysis. Like the play on its name, this course assumes no prior knowledge in Python language and aims to teach prospective students the basics on Python language structure AND the steps in analyzing real data. The course does not pretend that data analytics is easy and cut-corners to simplify anything. It is a very 'honest' demonstration that effectively gives overly ambitious future data analysts a flick on the forehead about data analysis. Who are we kidding? Data analysis using programming language requires sturdy knowledge in some nifty codes clean, splice and feature engineer the raw data and real critical thinking on figuring out 'Pythonic' ways to answer analytical questions. What does it even mean by Pythonic ways? Please refer to this article by Robert Clark, How to be Pythonic and Why You Should Care. We can discuss it somewhere down the line, when I am more experienced to understand it better. But for now, Packt Hub has the more comprehensive simple answer; it simply is an adjective coined to describe a way/code/structure of a code that utilizes or take advantage of the Python idioms well and displays the natural fluency in the language.

The bottom line is, we want to be able to fully utilize Python in its context and using its idioms to analyze data.

The course is conducted at Jovian.ai platform by its founder; Aakash and it takes advantage of Jupyter-like notebook format; Binder, in addition to making the synchronization available at Kaggle and Google's Colab. Each webinar in this course spans over close to 2 hours and each week, there are assignments on the lecture given. The assignments are due in a week but given the very disproportionate ratio of students and instructors, there were some extensions on the submission dates that I truly was grateful for. Forum for students is available at Jovian to engage students into discussing their ideas and question and the teaching body also conducts office hours where students can actively ask questions.

The instructor's method of teaching is something I believe to be effective for technical learners. In each lectures, he will be teaching the codes and module requires to execute certain tasks in the thorough procedure of the data analysis task itself. From importing the .csv formatted data into Python to establishing navigation to the data repository...from explaining what the hell loops are to touching base with creating functions. All in the controlled context of two most important module for the real objective of this course; Numpy and Pandas.

My gain from this course is immensely vast and that's why I truly think that freeCodeCamp and Jovian.ai really put the word 'tea' to 'teachers'. Taking advantage of the fact that people are involuntarily quarantined in their house, this course is something that should not be placed aside in the 'LATER' basket. I managed to clear my head to understand what 'loop' is! So I do think it can solve the world's problem!

In conclusion, this is the best course I have ever completed (90%!) on data analysis using Python. I look forward to attending it again and really finish up that last coursework.

Oh. Did I not mention why I got stuck? It was the last coursework. We are required to demonstrate all the steps of data analysis on data of our choice, create 5 questions and answer them using what we've learned throughout the course. Easy eh? Well, I've always had the tendency of digging my own grave everytime I get awesome cool assignments. But I'm not saying I did not do it :). Have a look-see at this notebook and consider the possibilities you can grasp after you've completed the course. And that's just my work...I'm a standard C-grade student.

And the exciting latest news from Jovian.ai is that they have upcoming course at Jovian for Deep Learning called Deep Learning with PyTorch: Zero to GANS! That's actually yesterday's news since they organized it earlier this year...so yeah...this is an impending second cohort! Tentatively, the course will start on Nov 14th. Click the link below to sign-up and get ready to attack the nitty-gritty. Don't say I didn't warn ya.

And that's me, reporting live from the confinement of COVID pandemic somewhere in a developing country at Southeast Asia....

🧰 Publicly available data

Hunting for spatial data comes naturally now. There seems to be less and less opportunity for doubts when we could attach a pair of coordinates to some places.

For work and hobby, hunting for data take almost half of the usable hours I set aside to execute certain objectives; if not 100%. Although the internet is a vast plain of data, not all of them are usable. The democratization of data is a subject that is to translucent to discuss but to solid to argue with. Thus, with differing opinions, we get different versions of them online. Here are some of the interesting data platforms I manage to scour based on their thematic subject

🌳 Nature and Environment

Delta at Risk - Profiling Risk and Sustainability of Coastal Deltas of the World. I found this while lamenting on how people love asking for data addition into their maps at the eleventh hour. I find their confidence in my skills quite misleading but flattering nonetheless. But it does not make it any less troublesome.

Protected Planet - Discover the world's protected and conserved areas. This platform includes not just data of protected areas, but also other effective area-based conservation measures like ICCAs IUCN listing and as the website claims, it is updated regular via submissions from agencies. So far, I found this platform to be the most convenient since it rounds up all possible conservation-based themes which also includes World Heritage Sites.

Global Forest Change (2000-2020) - The global forest extent change since 2000 to the current year or lovingly referred to as the Hansen data by most forestry RS specialist. This data is updated annually and to be honest, the platforms are literally everywhere. But this platform is legitimate under Earth Engine Apps and you can refer to Google Earth Engine for future data updates to ease your search.

👩‍⚖️ Administrative Data

GADM - Map and spatial data for all countries and their sub-divisions.

🏦 Built-environment Data

OpenStreet Map - This database is the most amazing feat of tech-aware crowdsourcing. A little more than 2 decades ago, some 'experienced' gate-keeping professionals would have refuted its legitimacy within an inch of their lives but OSM has proven that time prevails when it comes to bringing the accessibility and network data into practical use. I am not that adept with downloading from this website so I go directly to a more manual data download. My favorite is the Geofabrik Download but you can also try Planet OSM.

🎮 Other Cool Data

OpenCell ID - Open database platform of global cell towers. Cleaning the data is a nightmare but I think it is just me. I have little patience for cerebral stuff.

So, those are some of the data I managed to dig for personal projects. Hope it helps you guys too!

Story Map for Noobs: Cascade | WWF Network

Story Map is a web application template product that has been popularized in ArcGIS Online for a user-friendly and comprehensive narrative of maps. The ‘Cascade’ template has become the seamless interface of choice due to it’s ribbon transitions and availability of content streaming from external sources. 

Please refer to the following link for resources used in this webinar:

Story Map for Noobs: Cascade web application

📌 Availability: Retracted in 2021

The devil in the details

I have started to post some videos demonstrating some tools in ArcGIS Pro. Short ones and pretty quick ones which I strived for since I absolutely am frightened with the idea of irritating people with unnecessary voice-over. It has no garnered much response and it's cool with me. Although, the lack of traction does things to my insides, I go back to the real reason I am doing thing, which is to stash the tools that I managed to learn on my own by trials and errors and keep them somewhere I can refer back to it to remember how it works.

Creating maps involves a number of iterative processes made to suit the intended output. Although creating maps itself is a form of art; heavily reliant on target audience's knowledge and aesthetical preference, it is still an inherently democratic science. Thus, knowing the mainstream technology and tools in the industry to express your vision or message is given. So for those just starting out with using geographical information software (GIS) for your final year project or research, this videos are meant for you. The purpose is not to overwhelm you with too many information, or distract you with my narration, but to follow in real-time the process from the start up of the software to the running of tools that generates the information needed.

Knowing fully well that there is an endless variety of GIS software or tools out there, processes that you need to execute to make things happen may vary in name and functionalities. Forget the beef between ArcGIS and QGIS, of which one is the better tool; if it serves your needs, then use it. You're not obliged to pledge loyalty to software or brands although you are encouraged to maintain integrity in your beliefs when it comes to corporate versus open source tools in the industry. Both choices come with their advantages and disadvantages. Yours truly uses QGIS and ArcGIS Pro interchangeably. If it doesn't work in ArcGIS Pro, which I use primarily, I'll jump to using QGIS. It's not a big deal. If it works painlessly, there is no reason to feel bad about using it.

So far, the content I have made emphasizes mostly on ArcGIS Pro or Esri products since using them is how I come to learn more about geology and geography. QGIS was a name I did not learn of in my university years when ArcGIS versions start with the digit 9️⃣, so you can catch my drift.

We can go on and on about theoretical stuff and our smarter pals usually knows what to do when faced with the tools. Unfortunately, I fall in the percentile that needed to land on the job to understand what on earth I am supposed to do. This series of videos are for those who have the same problem as I do and need to see the magic actually happening before knowing what to do. And for the most part, there are so many things to read and try out before you get it right. So hopefully, the demos can kickstart some thoughts or observation in the logic within the software's ecosystem and become more than just a technical power-user.

This week, I touched on some tools that I found helpful when dealing with point vector data, so feel free to check it out 👇🏻

Next week, I'm thinking of exploring some series of point analysis and space time cube is beckoning for me to test it out. Until then, stay cool and drop a word if you need any clarifications on the demos!

[2022] 30 Day Map Challenge -- FAILED

Last year, I participated once again in the 30 Day Map Challenge that was going around in Twitter-ville come November. It is the 3rd attempt at the marathon and 2022 served as a reminder that progressed too despite getting stuck at Day 3 as life caught up with me.

I don't like the idea that I have left the challenge incomplete, again. It was not my priority and I work better with clear goals or visions of expected output. If it does not add to my need to learn something new ...it will be a task bound to head straight to the backburner. Let's resolve to make it a long-term routine instead of a spurt of stress trying to make the deadline.

As a consequence, I am attuning this task into one that actually gives me the benefit out putting into record the techniques and tools I used to make the maps in writing. I believe that will serve more purpose and added value other than visuals. And perhaps, have some stock ready for submission this year instead.

Anyone else participated in this challenge back in November? How did you do and what would you like to do better for the next one? Don't be shy and do drop a word or two.

Python: Geospatial Environment Setup (Part 2)

Python: Geospatial Environment Setup (Part 2)

Hey again folks! I am here for the second part of Python environmental setup for a geospatial workspace. I published the first part of this post two weeks ago. So if you've not yet read that, I'll catch you up to speed with our checklist:

Install Python ☑

Install Miniconda ☑

Install the basic Python libraries ☑

Create a new environment for your workspace

Install geospatial Python libraries

🗃 Create a new environment for your workspace

Since we have actually manually set up our base environment quite thoroughly with all the basic libraries needed, to make our work easier, we can just clone the base environment and install all the additional essential libraries needed for geospatial analysis. This new environment will be called geopy. Feel free to use a name you identify most with.

Why don't we just create a new environment? Well, it means we have to start installing the Python libraries again from scratch. Although it is no trouble to do so, we want to avoid installing so many libraries all at once. As I mentioned in Part 1, there is always a risk where incomplete dependencies in one library will affect the installation of other libraries that you intend to install in one go. Since we already have a stable and usable base environment, we can proceed to use it as a sort of pre-made skeleton that we will build our geospatial workspace with.

1️⃣ At the Anaconda Command Prompt, type the following:

2️⃣ Press Enter and the environment will be clone for you. Once it is done, you can use the following command to check the availability of your environment 👇🏻

You should be able to see your geopy environment listed along with the base environment.

👩🏻‍💻 Install geospatial Python libraries

Here we will proceed with the installation of a few geospatial Python libraries that are essential to reading and exploring the vectors and rasters.

🔺 fiona: This library is the core that some of the more updated libraries depend on. It is a simple and straightforward library that reads and writes spatial data in the common Python IOs without relying on the infamous GDAL's OGR classes.

🔺 shapely: shapely library features the capability to manipulate and edit spatial vector data in the planar geometric plane. It is one of the core libraries that recent geospatial Python libraries rely on to enable the reading and editing of vector data.

🔺 pyproj: is the Python interface for the cartographic projections and coordinate system libraries. Another main library that enables the 'location' characteristics in your spatial data to be read.

🔺 rasterio: reads and writes raster formats and provides a Python API based on Numpy N-dimensional arrays and GeoJSON.

🔺 geopandas: extends the pandas library to allow spatial operations on the geometric spatial data i.e shapefiles.

💀 As you might have noticed, we won't be doing any direct gdal library installation. It's mainly due to the fact that its installation is a process that seems to be accompanied by misery at every turn and involved workarounds that are pretty inconsistent for different individuals. Does it mean that we won't be using it for our Pythonic geospatial analysis? Heck no. But we will be taking advantage of the automatic dependency installation that comes with all the libraries above. The rasterio library depends on gdal and by installing it, we integrate the gdal library indirectly into our geospatial environment. I found that this method is the most fool-proof. Let's proceed to the installation of these libraries.

1️⃣ At the Anaconda Command Prompt, should you start from the beginning, ensure that your geopy environment is activated. If not, proceed to use the following command to activate geopy.

Once activated, we can install the libraries mentioned one after another. Nevertheless, you also have the option of installing them in one go directly using a single command 👇🏻

💀 geopandas is not included in this line-up NOT because we do not need it. It's another temperamental library that I prefer to isolate and install individually. If gdal is a rabid dog...then geopandas is a feral cat. You never know how-when-why it doesn't like you and forces a single 10-minute installation drag to hours.

3️⃣ Once you're done with installing the first line-up above, proceed with our feral cat below 👇🏻

4️⃣ Use the conda list command again to check if all the libraries have been installed successfully.

🎉Et voilá! Tahniah! You did it!🎉

🎯 The Jupyter Notebook

It should be the end of the road for the helluva task of creating the geospatial environment. But you're going to ask how to start using it anyway. To access this libraries and start analyzing, we can easily use the simple and straight-forward Jupyter Notebook. There are so many IDE choices out there but for data analysis, Jupyter Notebook suffices for me so far and if you are not familiar with Markdown, this tool will ease you into it slowly.

Jupyter Notebook can be installed in your geopy environment as follows:

And proceed to use it by prompting it open via the command prompt

It ain't that bad, right? If you're still having problems with the steps, do check out the real-time video I created to demonstrate the installation. And feel free to share with us what sort of problems you have encountered and the workaround or solutions you implemented! It's almost never a straight line with this, trust me. As mentioned in the previous post, check out the quick demo below 👇🏻

See you guys again for another session on geospatial Python soon!