- 2023.8.14: Updated code
- Install PyTorch
conda install -c conda-forge imageio h5py
Available config registries
{'mini-imagenet_basic': fewshot.configs.mini_imagenet_config.BasicConfig,
'mini-imagenet_kmeans-refine': fewshot.configs.mini_imagenet_config.KMeansRefineConfig,
'mini-imagenet_kmeans-distractor': fewshot.configs.mini_imagenet_config.KMeansDistractorConfig,
'mini-imagenet_imp': fewshot.configs.mini_imagenet_config.ImpModelConfig,
'mini-imagenet_crp': fewshot.configs.mini_imagenet_config.CRPConfig,
'mini-imagenet_map-dp': fewshot.configs.mini_imagenet_config.MapDPConfig,
'mini-imagenet_soft-nn': fewshot.configs.mini_imagenet_config.SoftNNConfig,
'mini-imagenet_dp-means-hard': fewshot.configs.mini_imagenet_config.DPMeansHardConfig,
'omniglot_basic': fewshot.configs.omniglot_config.BasicConfig,
'omniglot_kmeans-refine': fewshot.configs.omniglot_config.KMeansRefineConfig,
'omniglot_kmeans-distractor': fewshot.configs.omniglot_config.KMeansDistractorConfig,
'omniglot_imp': fewshot.configs.omniglot_config.ImpModelConfig,
'omniglot_crp': fewshot.configs.omniglot_config.CRPConfig,
'omniglot_map-dp': fewshot.configs.omniglot_config.MapDPConfig,
'omniglot_soft-nn': fewshot.configs.omniglot_config.SoftNNConfig,
'omniglot_dp-means-hard': fewshot.configs.omniglot_config.DPMeansHardConfig,
'tiered-imagenet_basic': fewshot.configs.tiered_imagenet_config.BasicConfig,
'tiered-imagenet_kmeans-refine': fewshot.configs.tiered_imagenet_config.KMeansRefineConfig,
'tiered-imagenet_kmeans-distractor': fewshot.configs.tiered_imagenet_config.KMeansDistractorConfig,
'tiered-imagenet_imp': fewshot.configs.tiered_imagenet_config.ImpModelConfig,
'tiered-imagenet_crp': fewshot.configs.tiered_imagenet_config.CRPConfig,
'tiered-imagenet_map-dp': fewshot.configs.tiered_imagenet_config.MapDPConfig,
'tiered-imagenet_soft-nn': fewshot.configs.tiered_imagenet_config.SoftNNConfig,
'tiered-imagenet_dp-means-hard': fewshot.configs.tiered_imagenet_config.DPMeansHardConfig}
This repository is adapted from https://github.com/renmengye/few-shot-ssl-public for PyTorch 0.3.1
submit_omniglot.sh provides example usage of the main file.
We also have submission scripts for running code on a slurm cluster.
Please refer to submit_all_models.sh and submit_super.sh for examples.