---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: label
    dtype:
      class_label:
        names:
          '0': Amrapali
          '1': Banana
          '2': Bari 4
          '3': Fazli
          '4': GobindoBhog
          '5': GopalBhog
          '6': Harivanga
          '7': Himsagar
          '8': Khrishapat
          '9': Langra
          '10': RaniBhog
          '11': Sundari
  splits:
  - name: train
    num_bytes: 17855503855
    num_examples: 3900
  download_size: 18516857667
  dataset_size: 17855503855
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
license: cc-by-4.0
task_categories:
- image-classification
size_categories:
- 1K<n<10K
---

# MangoClassify 12 Variety Classification

A dataset for variety classification of common mangoes. The dataset contains 3,900 images across 12 classes:  
Images per class:
- Amrapali: 600
- Banana: 212
- Bari 4: 240
- Fazli: 120
- GobindoBhog: 41
- GopalBhog: 406
- Harivanga: 575
- Himsagar: 502
- Khrishapat: 380
- Langra: 506
- RaniBhog: 92
- Sundari: 226

This dataset is indexed on https://project-agml.github.io/ as part of the AgML python library.

## Citation

```bibtex
@article{rahman2025mangoclassify,
  title={MangoClassify-12: A high-resolution image dataset of twelve indigenous Bangladeshi mango cultivars},
  author={Rahman, Md Sajedur and Nahin, Md Mahfuz Ahmed and Rahman, Md Mahbubur and Rani, Mollika and Islam, Md Ashraful and Bashir, Al and Shafkat, Ahmad and Mallik, Bijon and Majeed, Yaqoob},
  journal={Data in Brief},
  pages={112037},
  year={2025},
  publisher={Elsevier}
}
```

Md. Sajedur Rahman, Md. Mahfuz Ahmed Nahin, Mollika Rani, and MD Ashraful Islam. (2025). MangoClassify-12: Native Mango Dataset from BD [Dataset]. Kaggle. https://doi.org/10.34740/KAGGLE/DSV/12460544