FC100 (Fewshot-CIFAR100)

Introduced by Oreshkin et al. in TADAM: Task dependent adaptive metric for improved few-shot learning

The FC100 dataset (Fewshot-CIFAR100) is a newly split dataset based on CIFAR-100 for few-shot learning. It contains 20 high-level categories which are divided into 12, 4, 4 categories for training, validation and test. There are 60, 20, 20 low-level classes in the corresponding split containing 600 images of size 32 × 32 per class. Smaller image size makes it more challenging for few-shot learning.

Source: Prototype Rectification for Few-Shot Learning

Homepage

Benchmarks

Add a new result Link an existing benchmark

Task	Dataset Variant	Best Model
Few-Shot Image Classification	FC100 5-way (1-shot)	BAVARDAGE
Few-Shot Image Classification	FC100 5-way (5-shot)	BAVARDAGE
Few-Shot Image Classification	Fewshot-CIFAR100 - 1-Shot Learning	pseudo-shots
Few-Shot Image Classification	Fewshot-CIFAR100 - 5-Shot Learning	pseudo-shots
Few-Shot Image Classification	FC100 5-way (10-shot)	MTL