Pet Image Classification Dataset With Three Animal Classes

Mini classification dataset containing images of cats, dogs, and rabbits for small-scale projects. The dataset was created by author 'rokmr' and was last updated on Hugging Face in July 2023. It is designed for the task of predicting pet type from an image.

Use Cases

• Train a convolutional neural network to classify images as 'cat', 'dog', or 'rabbit' using the provided integer labels.
• Benchmark lightweight model architectures on a three-class image classification task.
• Fine-tune a pre-trained vision model on the specific domain of common household pets.
• Test data augmentation strategies for improving model performance on animal image data.

Strengths

• Contains three distinct, common pet classes for a clear classification task.
• Described as 'mini' with a size of 24 MB, suitable for quick experimentation.

Limitations

• Exact row count is unknown, making it difficult to assess dataset scale.
• Potential geographic or breed bias is unknown as image sources are not specified.
• Limited to only three animal classes, excluding other common pets.

Provenance

Source

Hugging Face dataset uploaded by user 'rokmr'.

Collection Method

null

Time Range

null

Freshness

Last updated on 2023-07-25.

Geography

null