--- # For reference on dataset card metadata, see the spec: # https://github.com/huggingface/hub-docs/blob/main/datasetcard.md?plain=1 # Doc / guide: # https://huggingface.co/docs/hub/datasets-cards pretty_name: "AuraClima" tags: - agriculture - climate - co2-emissions - time-series - forecasting - machine-learning - deep-learning - lstm - environmental-data task_categories: - tabular-regression - time-series-forecasting license: other --- # Dataset Card for AuraClima AuraClima is a collection of agricultural and climate-related datasets, processed data, trained machine learning models, and preprocessing artifacts developed to support the AuraClima application. The repository brings together agriculture and food-related CO₂ emissions data, historical CO₂ emissions data, processed datasets, trained Keras models, and supporting preprocessing artifacts used by the application. The dataset repository is intended to provide the data and machine learning artifacts required by the AuraClima application while also making these resources available for experimentation, research, and further development. ## Dataset Details ### Dataset Description AuraClima combines agricultural and climate-related data with machine learning and deep learning artifacts. The repository currently contains: - 4 CSV datasets - 3 trained Keras models - 1 feature-definition file - 4 serialized preprocessing/scaling artifacts The datasets include agricultural and food-related CO₂ emissions data, historical CO₂ emissions data covering 1960–2018, cleaned agricultural data, and a MinMax-scaled CO₂ emissions dataset. The repository also includes trained models for agriculture-related prediction, agricultural time-series modeling, and CO₂ emissions forecasting. - **Curated by:** Abdullah Imran - **Funded by:** Not specified - **Shared by:** Abdullah Imran - **Language(s) (NLP):** Not applicable - **License:** Not specified ### Dataset Sources - **Dataset Repository:** https://huggingface.co/datasets/AbdullahImran/AuraClima - **Application / Demo:** https://huggingface.co/spaces/AbdullahImran/AuraClima - **Paper:** Not specified ## Uses ### Direct Use AuraClima can be used for experimentation and research involving: - Agricultural data analysis - CO₂ emissions analysis - Climate-related data analysis - Time-series forecasting - Machine learning - Deep learning - LSTM-based forecasting - Agricultural prediction - Environmental data analysis - Development and evaluation of predictive models The repository also contains trained Keras models that can be used for experimentation and further model development. ### Out-of-Scope Use The dataset and included models should not be used as the sole basis for: - Climate or environmental policy decisions - High-stakes environmental decisions - Safety-critical applications - Unvalidated production systems - Scientific claims about future climate conditions without independent validation - Predictions outside the distribution of the data without appropriate evaluation Model predictions should not be interpreted as authoritative climate forecasts without appropriate scientific validation. ## Dataset Structure The repository contains 12 files with an approximate local size of 19 MB. ```text AuraClima/ ├── Agri_Slider_Model.keras ├── Agri_TimeSeries.keras ├── Agrofood_co2_emission.csv ├── cleaned_agriculture_data.csv ├── CO2_Emissions_1960-2018.csv ├── CO2_Emissions_MinMaxScaled.csv ├── co2_lstm_forecast_model.keras ├── feature_cols2.list ├── scaler1.save ├── scaler3.save ├── scalerX2.save └── scalerY2.save