Dataset Viewer

The dataset viewer is not available for this subset.

Cannot get the split names for the config 'default' of the dataset.

Exception:    SplitsNotFoundError
Message:      The split names could not be parsed from the dataset config.
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.12/site-packages/datasets/inspect.py", line 286, in get_dataset_config_info
                  for split_generator in builder._split_generators(
                                         ^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 82, in _split_generators
                  raise ValueError(
              ValueError: The TAR archives of the dataset should be in WebDataset format, but the files in the archive don't share the same prefix or the same types.
              
              The above exception was the direct cause of the following exception:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/split_names.py", line 65, in compute_split_names_from_streaming_response
                  for split in get_dataset_split_names(
                               ^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/inspect.py", line 340, in get_dataset_split_names
                  info = get_dataset_config_info(
                         ^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/inspect.py", line 291, in get_dataset_config_info
                  raise SplitsNotFoundError("The split names could not be parsed from the dataset config.") from err
              datasets.inspect.SplitsNotFoundError: The split names could not be parsed from the dataset config.

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

YAML Metadata Warning: empty or missing yaml metadata in repo card

Check out the documentation for more information.

Latent GIG Backup

Research project backup for Manifold-Aware Gradient Integration methods - exploring various path-based explainability methods for image classifiers including latent space paths, spectral decomposition, and manifold-aware approaches.

Download & Extract

# huggingface_hub 설치 (필요시)
pip install huggingface-hub==0.20.2

# 다운로드
huggingface-cli download leekwoon/260204_magig_backup --repo-type dataset --local-dir ./latent_gig_data

# 무결성 확인 (선택사항)
cd latent_gig_data
md5sum -c checksums.md5

# 파일 합치기 및 압축 해제
cat data.tar.gz.part_* | tar -xzvf -

Directory Structure

latent_gig/
├── cleanig/              # Core implementation modules
│   ├── explainer/        # Explainability algorithms (40+ methods)
│   ├── dataset/          # Dataset utilities (ImageNet, OxfordPet, CIFAR10)
│   ├── classifier/       # Classifier utilities
│   ├── metric/           # Evaluation metrics (DiffID, ROAR)
│   ├── mar_vae/          # MAR-VAE implementation
│   └── vae_wrapper.py    # VAE wrapper for different models
├── configs/              # Hydra configuration files (40+ configs)
├── guided-diffusion/     # Guided diffusion model
├── models/               # Pre-trained models
├── notebooks/            # Jupyter notebooks for analysis
├── results/              # Experiment results
│   ├── benchmark_diffid/ # DiffID benchmark results
│   ├── roar/             # ROAR evaluation results
│   └── classifier_*/     # Trained classifiers
├── scripts/              # Execution scripts
│   ├── icml26_*/         # ICML 2026 experiment scripts
│   ├── diffid.py         # DiffID evaluation
│   ├── roar.py           # ROAR evaluation
│   └── measure_computation_time.py
├── papers/               # Related papers
└── requirements.txt      # Python dependencies

Key Components

Explainers (40+ methods)

Baseline Methods:

IG: Integrated Gradients (standard linear path)
GIG: Guided IG (gradient-guided adaptive paths)
AGI: Adversarial Gradient Integration
BIG: Blur Integrated Gradients
SAMP: Adaptive sampling method
XRAI: Region-based attribution

Latent Space Methods:

Latent GIG: Latent space paths with VAE models
EIG: Enhanced IG in latent space
MIG: Manifold IG (geodesic paths on VAE manifold)
Spectral EIG: Spectral decomposition in latent space

Spectral Methods:

Spectral IG: SVD-based coarse-to-fine path construction
Multiscale Spectral IG: Multi-resolution spectral paths

Experimental Methods:

MCTS GIG: Monte Carlo Tree Search guided paths
Curvature GIG: Curvature-aware path construction
Boundary IG: Decision boundary aware paths

VAE Models Supported

MAR: Masked Auto-Regressive VAE
SD1/SD2: Stable Diffusion VAE (v1, v2)
KD: Kandinsky VAE

Datasets

ImageNet (ILSVRC2012)
Oxford-IIIT Pet Dataset
CIFAR-10

Evaluation Metrics

DiffID: Insertion/Deletion curves
ROAR: RemOve And Retrain
Computation time benchmarks

Usage

# Install
pip install -e .

# Run evaluation
python scripts/diffid.py --config-name latent_gig dataset=imagenet
python scripts/roar.py --config-name spectral_ig dataset=oxfordpet

# Measure computation time
python scripts/measure_computation_time.py --config-name ig dataset=imagenet

Citation

Research project for ICML 2026 submission on manifold-aware gradient integration methods for explainable AI.

Downloads last month: 8