The dataset includes both train,validation and test images, making it suitable for use in training and evaluating machine learning models for disease diagnosis. (tested on YOLOv3 and YOLOv4).
The dataset is composed of a total of 120 images, where 6 is the test images, of which 98 images are used for training and 16 images are used for validation. The images have been labeled with one of the following disease classes: bacteria, brown_spot, leaf_smut. Each image is a 416x416 RGB image of a rice leaf. The dataset also includes a CSV and json file with additional metadata such as the date the image was taken and the location where the image was taken.
you can find the full dataset from : https://archive.ics.uci.edu/ml/datasets/rice+leaf+diseases