Dark Mode
Climate Variable Classification Data
Data Science and Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This collection of data is designed for predicting weather conditions. It provides detailed meteorological measurements, enabling users to act as weather experts and train machine learning models to forecast future weather events. The primary objective is to use the numerous input features to predict the weather description, which serves as the designated output column for classification tasks.
Columns
The dataset contains 34 columns, all of which are 100% valid with no missing values. Key columns include:
- timestamp_utc: The timestamp of the weather forecast in Coordinated Universal Time.
- temp: The observed temperature, with values ranging from 21.9 to 50.2 (Mean: 37.2).
- app_temp: The apparent temperature (ranging from 18.7 to 50.2).
- precip: Precipitation measurements in millimeters (Mean: 0, Max: 0.04).
- ghi: Global Horizontal Irradiance.
- dhi: Diffuse Horizontal Irradiance.
- dni: Direct Normal Irradiance.
- rh: Relative humidity (Mean: 69.8, Max: 100).
- wind_spd: Wind speed (Mean: 3.79).
- wind_cdir: Categorical direction of wind (e.g., NNW is the most common at 30%).
- slp/pres: Sea level pressure and atmospheric pressure readings.
- clouds: Total cloud cover, split into high, mid, and low layers (e.g.,
clouds_hi,clouds_mid,clouds_low). - ozone: Ozone levels (Mean: 319).
- description: The target output column, containing textual descriptions of the weather (e.g., 'Few clouds' is the most common description at 45%).
- snow/snow_depth: Both snow and snow depth columns show a mean of 0, indicating zero snowfall across the sampled period.
Distribution
The data is provided in a tabular format, typically stored as a CSV file named
weather_data.csv. The file size is 7.79 kB. It is structured with 34 columns and contains 40 records or rows, all of which are noted as 100% valid, indicating excellent data quality with no missing or mismatched entries in this sample.Usage
This data is highly suitable for several applications:
- Training multi-class classification models to predict the weather description based on physical parameters.
- Developing and testing predictive algorithms for short-term weather forecasting.
- Analysing correlations between various atmospheric parameters (irradiance, temperature, pressure) and resulting weather phenomena.
- Projects focused on intermediate-level machine learning and data analysis in climate science.
Coverage
The temporal scope of the forecast data runs from 2025-01-15 to 2025-01-20, covering a short, six-day period. Both UTC and local timestamps are provided. The geographic scope is not specified within the available metadata. The dataset is labelled as 'Never' for expected update frequency.
License
CC BY-NC-SA 4.0
Who Can Use It
Intended users include data scientists building predictive models, meteorological researchers studying atmospheric conditions, and machine learning practitioners seeking high-quality, fully validated tabular data for intermediate classification tasks. It is particularly useful for those seeking to become weather experts by analysing the data.
Dataset Name Suggestions
- Short-Range Weather Forecast Parameters
- Meteorological Prediction Data Sample
- Climate Variable Classification Data
- January 2025 Weather Observations
Attributes
Original Data Source: Climate Variable Classification Data
QUALITYLoading...