HAD TO REDO THE README.md

Well… I thought I was finally done.
SOO now Welcome to the unexpected devlog # 7

Turns out my previous README had a bit too much AI in it, so I decided to rewrite almost the entire thing myself. It definitely took way longer than I expected, but now it actually feels like my project.

At this point everything is finally done:
✅ UI
✅ Graphs
✅ Deployment
✅ Comments
✅ README

Now hoping to not write another devlog now
Byeeee!!!!

I DID THE FINAL STEPS

Well almost all of it was done the remaining things were

• README.md file (half well)
• Comment of the entire files (I don’t like to do commenting 😭😭😭😭😭😭)

WELLL WELLL
Funny enough, commenting turned out to be the easier part. The real challenge was writing the README because I wanted it to actually explain the project properly instead of being just a few lines.

End of devlog # 6
NOW GONNA SHIP THE PROJECT
Everything is now ready… all that’s left is to hit Ship Project 🚀

UI UI UI ALL I DID WAS UI !!!!!!

This update has been all about polishing the project.

I’ve finally completed the UI for every single page, you all can see the before and the after in the screenshots of the UI.

Besides this I had successfully deployed the application! despite the errors I faced while making it run smooth 😅

Looking back, it’s hard to believe this started as my first Streamlit project. I spent way more time tweaking layouts, colors, containers, and spacing than I ever expected, but it was worth it FINALLY!!!!

Now the dashboard finally feels complete, and all that’s left is the final polish: cleaning up the code, adding comments, and finishing the README before I officially ship it.

🌍 The app is now live! If you’d like to explore 80+ years of climate data through interactive visualizations, give it a try:

THE LINK TO THE PROJECT:

https://the-earth-kept-a-record-here-it-is-vavg5teg69z5fpeqsqcren.streamlit.app/

Everyone do let me know about any improvements that you may seem that I can make before the shipping of this project

This is the end of devlog # 5
See you all in devlog # 6

Byeeeeeee!

WHY DOES IT LOOK LIKE THAT?????????

So, I was finally working on my last graph, the one that predicts future temperature anomalies for a city or country.

I built the model, generated the predictions… and this is what I got. (the graph with almost a straight line)

Honestly, I was disappointed. I expected something that looked more like a “prediction,” not just a line.

So, I tried connecting the future with the historical data, changed a few things, and even wondered if I had done Polynomial Regression completely wrong.

Turns out… the model wasn’t the problem.

Regression doesn’t make random ups and downs, it simply learns the overall trend from the historical data and continues it into the future.

In the end, I stopped trying to make the graph look dramatic and focused on making it accurate instead.

What do you all think of this graph??

This is the end of devlog #4 

See you all in devlog #5
Byeeeee

🚨 I USED THE WRONG DATA ALL ALONG…

So Storytime!!!!

When I almost like coded 3 out of 5 graphs, only then did I realise that I had the MAIN ingredient wrong: the data.

To be exact, when I first downloaded the datasets, they seemed to have everything I needed:

• 🌡️ Temperatures from the 1940s to 2025
• 📅 Year-wise timeline
• 🌍 Data for cities and countries
• ❓Temperature anomaly (wait… what is this?)

At first, I honestly thought Temperature Anomaly  was just a fancy way of saying temperature change. Turns out I need to work a bit harder on both my SAT English and this project. 😅 Because when I looked into the csv in detail the Temperature anomaly was like in range of   -0.000342 to 4.66247 and this was for like every single country each varing differently.

After a bit of Googling and reading, I discovered that ** temperature anomaly is not the actual temperature**. Instead, it measures how much warmer or cooler a place is compared to its average temperature during a reference period.

That meant I had a huge problem.

My country data  was already stored as temperature anomalies, but my city data contained actual temperatures. I was unknowingly comparing two completely different measurements in the same project.

If I had continued like that, my comparisons would have been misleading, my rankings would have been unfair, and my prediction model would have been built on inconsistent data.

So… I stopped everything.

I went back, learned how temperature anomalies are calculated, converted all of my city datasets to use the same measurement, and rebuilt the affected parts of the project. It definitely wasn’t the plan, but I’m glad I caught the mistake before shipping the project.

End of story…….

Besides this I improved my horrible looking homepage (the simple screenshot) and now it definitely looks like something to look at (the attractive homepage) this also had me going through streamlit’s documentation and continuous learning……
which I am getting fed up of but again that is what coding means right?😅

Lastly I did also complete my 4 out of 5 graphs which is definitely a relief (pheww) and now all that’s left todo is 

• Graph 5
• Improved UI of all the graph pages
• About page

Girls and Guys I  would love to have your feedback for my project

(PLZ PLZ PLZ suggest improvements 😭😭😭😭)

Byeeee!!!!!
See you in Devlog # 4 ✨

END of devlog # 3

🚨 I USED THE WRONG DATA ALL ALONG…

So Storytime!!!!

When I almost like coded 3 out of 5 graphs, only then did I realise that I had the MAIN ingredient wrong: the data.

To be exact, when I first downloaded the datasets, they seemed to have everything I needed:

• 🌡️ Temperatures from the 1940s to 2025
• 📅 Year-wise timeline
• 🌍 Data for cities and countries
• ❓Temperature anomaly (wait… what is this?)

At first, I honestly thought Temperature Anomaly  was just a fancy way of saying temperature change. Turns out I need to work a bit harder on both my SAT English and this project. 😅 Because when I looked into the csv in detail the Temperature anomaly was like in range of   -0.000342 to 4.66247 and this was for like every single country each varing differently.

After a bit of Googling and reading, I discovered that ** temperature anomaly is not the actual temperature**. Instead, it measures how much warmer or cooler a place is compared to its average temperature during a reference period.

That meant I had a huge problem.

My country data  was already stored as temperature anomalies, but my city data contained actual temperatures. I was unknowingly comparing two completely different measurements in the same project.

If I had continued like that, my comparisons would have been misleading, my rankings would have been unfair, and my prediction model would have been built on inconsistent data.

So… I stopped everything.

I went back, learned how temperature anomalies are calculated, converted all of my city datasets to use the same measurement, and rebuilt the affected parts of the project. It definitely wasn’t the plan, but I’m glad I caught the mistake before shipping the project.

End of story…….

Besides this I improved my horrible looking homepage (the simple screenshot) and now it definitely looks like something to look at (the attractive homepage) this also had me going through streamlit’s documentation and continuous learning……
which I am getting fed up of but again that is what coding means right?😅

Lastly I did also complete my 4 out of 5 graphs which is definitely a relief (pheww) and now all that’s left todo is 

• Graph 5
• Improved UI of all the graph pages
• About page

Girls and Guys I  would love to have your feedback for my project

(PLZ PLZ PLZ suggest improvements 😭😭😭😭)

Byeeee!!!!!
See you in Devlog # 4 ✨

END of devlog # 3

Devlog # 2

Hi everyone! 
Hope you’re all doing well! 😊

Thankfully, I got out of the data cleaning phase. This week,  I had my mind on the design and page layout of my web application, as well as the visualisations that will bring my idea to life.

🖥️ Planning the Website
I thought my web application would have a structure like this:

• A Homepage
• 5 pages,  each having a different graph type for each country
• An About page

But everyone, the main challenge was that I had never worked with streamlit so in short, I had to learn it, and for me, tutorials almost never worked for me, so I had to go through documentation every single time, and this got me a lot of time spent on it instead of code.

🎨 Building the Homepage

As I was trying to set up my homepage, I wanted something like this (this picture of the homepage is the inspo from ChatGPT)

So I started with the title and the choices, but when I wanted to take the user to that specific page of the graph, I got this error ( the picture with the error message). This had me stuck for almost an hour or so until I got it right, and FINALLY, my homepage was done. ( the homepage in black with the text “Earth Kept a record(in red), Here it is!!). Seeing this, **I was in the air!!!! YAAAAY!!!🎉🎉 **

📈 First Graph first try

I also started working on my first visualisation: the Temperature Trends Graph. This graph enables users to select a city or country and view how temperatures have evolved, utilising real climate data.To make it more visually appealing, I added a color-gradient effect to the graph. I did get some help from AI while experimenting with the design, but I’m really happy with how it turned out. It’s the first graph of the project, and I am gonna still work on it, and this progress makes me feel really GOOD.. 🌎(finally YAAAY YAAAY🎉🎉🎉)

✨That’s all for this week’s progress!

See you in Devlog #3 🚀 

Devlog # 2

Hi everyone! 
Hope you’re all doing well! 😊

Thankfully, I got out of the data cleaning phase. This week,  I had my mind on the design and page layout of my web application, as well as the visualisations that will bring my idea to life.

🖥️ Planning the Website
I thought my web application would have a structure like this:

• A Homepage
• 5 pages,  each having a different graph type for each country
• An About page

But everyone, the main challenge was that I had never worked with streamlit so in short, I had to learn it, and for me, tutorials almost never worked for me, so I had to go through documentation every single time, and this got me a lot of time spent on it instead of code.

🎨 Building the Homepage

As I was trying to set up my homepage, I wanted something like this (this picture of the homepage is the inspo from ChatGPT)

So I started with the title and the choices, but when I wanted to take the user to that specific page of the graph, I got this error ( the picture with the error message). This had me stuck for almost an hour or so until I got it right, and FINALLY, my homepage was done. ( the homepage in black with the text “Earth Kept a record(in red), Here it is!!). Seeing this, **I was in the air!!!! YAAAAY!!!🎉🎉 **

📈 First Graph first try

I also started working on my first visualisation: the Temperature Trends Graph. This graph enables users to select a city or country and view how temperatures have evolved, utilising real climate data.To make it more visually appealing, I added a color-gradient effect to the graph. I did get some help from AI while experimenting with the design, but I’m really happy with how it turned out. It’s the first graph of the project, and I am gonna still work on it, and this progress makes me feel really GOOD.. 🌎(finally YAAAY YAAAY🎉🎉🎉)

✨That’s all for this week’s progress!

See you in Devlog #3 🚀 

Devlog #1 🌍
Hi everyone!

I’m Shehma Ali, and welcome to my first devlog.

Today I’ll be sharing what my project is and what I’ve done so far.🚀

•My Project Visualizes 80 years of climate change using real NASA and Copernicus climate data.

•Lets users explore temperature trends for 10 cities from the 1940s to 2025.

•Includes city trend graphs, global comparisons, warming rankings, and heatmaps.

•Features future temperature predictions using linear regression with adjustable controls.

•Built entirely in Python using Pandas, Plotly, and Streamlit.

•Makes complex climate data simple and easy to understand.

🔨 Progress So Far

To be completely honest, the data collection stage has been brutal.

I first had to find city-level climate data from NASA and country-level climate data from Copernicus, which took a lot more searching than I expected. After gathering all the CSV files, I aligned them to the same year range and cleaned them so they could be used together in the project.

With the datasets now prepared FINALLY, I’m ready to start building the visualisations and bringing the app to life.
See you in the next devlog! 🌎✨