‘Bout the Task

You’re tasked with buildin’ a service that fetches weather info from the internet and presents it in a useful manner. The task is deliberately open-ended – you choose yer own technology, platform, and approach.

🌤️ Main Task

🎯 Målsetting

Målet med denne oppgaven er å utvikle en modell for å forutsi prisen på boliger basert på ulike faktorer. Modellen skal kunne brukes til å gi en indikasjon på hva en bolig kan være verdt, og kan være nyttig for både selgere og kjøpere.

🛠️ Datagrunnlag

Datagrunnlaget består av informasjon om boliger solgt i Oslo i perioden 2018-2023. Dataene inkluderer blant annet:

• Areal (m²)
• Antall soverom
• Beliggenhet (bydel)
• Byggeår
• Kjøpesum (NOK)

⚙️ Metode

Vi vil benytte oss av maskinlæring for å utvikle modellen. Vi vil eksperimentere med ulike algoritmer, som for eksempel lineær regresjon, beslutningstrær og tilfeldig skog. Modellen vil bli trent på en del av datagrunnlaget, og testet på en annen del for å evaluere ytelsen.

📈 Evaluering

Modellen vil bli evaluert ved hjelp av ulike metrikker, som for eksempel R-kvadrert, gjennomsnittlig absolutt feil (MAE) og rotmiddelkvadratfeil (RMSE). Vi vil også vurdere modellens evne til å generalisere til nye data.

🎯 Goal

The goal of this task is to develop a model to predict the price of homes based on various factors. The model should be able to provide an indication of what a home might be worth, and can be useful for both sellers and buyers.

🛠️ Data Basis

The data basis consists of information about homes sold in Oslo in the period 2018-2023. The data includes, among other things:

• Area (m²)
• Number of bedrooms
• Location (district)
• Year of construction
• Purchase sum (NOK)

⚙️ Method

We will use machine learning to develop the model. We will experiment with different algorithms, such as linear regression, decision trees, and random forest. The model will be trained on a part of the data basis, and tested on another part to evaluate performance.

📈 Evaluation

The model will be evaluated using various metrics, such as R-squared, mean absolute error (MAE), and root mean squared error (RMSE). We will also assess the model’s ability to generalize to new data.

Part 1 – Gatherin’ Weather Data

Yer service needs to fetch weather data for a specified location. You go on and figure out which source ya wanna use, and how to get structured data from it programmatically.

Part 2 – Show ‘em the Data

Once ya done wranglin’ up the data, ya gotta present it in a clear and easy-to-use manner.

• Think on what the user needs to know, plain and simple.
• The presentation gotta be tidy – whether it’s a dashboard, a table, a view in the terminal, or somethin’ else.

⭐ Part 1 – Smart Advice

Expand yer service to answer questions based on the weather data.

Examples of questions the service can answer:

• Do I need an umbrella?
• Do I need a jacket?
• Should I have winter tires?
• Do I need sunscreen?
• Is it slick out there?

These answers should be based on actual values from the weather data – not just static text.

⭐ Part 2 – Automatic Location Detectin’

If the user don’t specify a location, the service should automatically figure out where it’s runnin’ and use that location.

• Figure out how ya can determine position without the user writin’ nothin’
• This here should work as a “fallback” – the location can still be overridden manually