This dataset contains pictures of 50 different car parts. It’s divided into three sets: a training set, a testing set, and a validation set. Each set has images of all 50 car parts.
The training set doesn’t have an equal number of images for each part. The class “Ignition Coil” has the most training images, which is 200, while the class “Leaf Spring” has the fewest, only 110 images.
Both the validation and test sets have 5 images for each of the 50 parts. The images are all in JPG format and have dimensions of 224 x 224 pixels with 3 color channels (RGB). These are the original images, without any modifications or changes made to them.