README.md

---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: objects
    struct:
    - name: bbox
      list:
        list: float64
    - name: categories
      list:
        class_label:
          names:
            '0': grape
  splits:
  - name: train
    num_bytes: 913415542
    num_examples: 249
  download_size: 913427209
  dataset_size: 913415542
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---
# Embrapa Wgsid Grape Detection

A dataset for object detection of grape clusters. The dataset contains 249 images with 4,775 bounding box annotations across 1 category.

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