Overview

In addition to the tabular benchmark datasets provided in classical directory. We provide more details on the generation of the 10 CV/NLP datasets as below.

In a nutshell, pretrained models are applied to extract data embedding from CV and NLP datasets to access more complex representations, which has been widely used in AD literature [1,2,3]. Please refer to Section 3.2 and Appx. B.1 for details.

Adapted CV and NLP Datasets for Tabular AD

For the CV datasets, we provide two versions of transformed embedding data for evaluating tabular-based AD algorithms (codes are available in colab):

• ImageNet-pretrained ResNet-18 to extract the embedding after the last average pooling layer. See Data Folder.
• ImageNet-pretrained ViT to extract the embedding after the last average pooling layer. See Data Folder.

For the NLP datasets, we provide two versions of transformed embedding data (codes are available in colab)

• Pretrained BERT to extract the embedding of the [CLS] token. See Data Folder.
• Pretrained RoBERTa to extract the embedding of the [CLS] token. See Data Folder.

Raw Data File Download

CV datasets:

• CIFAR10: https://pytorch.org/vision/stable/generated/torchvision.datasets.CIFAR10.html
• FashionMNIST: https://pytorch.org/vision/stable/generated/torchvision.datasets.FashionMNIST.html#torchvision.datasets.FashionMNIST
• MNIST-C: https://zenodo.org/record/3239543
• MVTec-AD: https://www.mvtec.com/company/research/datasets/mvtec-ad
• SVHN: https://pytorch.org/vision/stable/generated/torchvision.datasets.SVHN.html#torchvision.datasets.SVHN

NLP datasets:

• Agnews: https://huggingface.co/datasets/ag_news
• Amazon: https://huggingface.co/datasets/amazon_polarity
• Imdb: https://huggingface.co/datasets/imdb
• Yelp: https://huggingface.co/datasets/yelp_review_full
• 20newsgroups: http://qwone.com/~jason/20Newsgroups/

---

Here we demonstrate the MVTec-AD data embedded by the pretrained ResNet-18 and ViT, respectively. These transformed (from image/text to tabular) data could be considered as a good baseline for evaluating different AD algorithms.

• MVTec-AD embedded by the pretrained ResNet-18
• MVTec-AD embedded by the pretrained ViT

How to Read Data?

We have unified all the datasets in .npz format, and you can read a dataset by the following script

import numpy as np
data = np.load('6_cardio.npz', allow_pickle=True)
X, y = data['X'], data['y']

Full Dataset List (57 datasets)

Number	Data	# Samples	# Features	# Anomaly	% Anomaly	Category
1	ALOI	49534	27	1508	3.04	Image
2	annthyroid	7200	6	534	7.42	Healthcare
3	backdoor	95329	196	2329	2.44	Network
4	breastw	683	9	239	34.99	Healthcare
5	campaign	41188	62	4640	11.27	Finance
6	cardio	1831	21	176	9.61	Healthcare
7	Cardiotocography	2114	21	466	22.04	Healthcare
8	celeba	202599	39	4547	2.24	Image
9	census	299285	500	18568	6.20	Sociology
10	cover	286048	10	2747	0.96	Botany
11	donors	619326	10	36710	5.93	Sociology
12	fault	1941	27	673	34.67	Physical
13	fraud	284807	29	492	0.17	Finance
14	glass	214	7	9	4.21	Forensic
15	Hepatitis	80	19	13	16.25	Healthcare
16	http	567498	3	2211	0.39	Web
17	InternetAds	1966	1555	368	18.72	Image
18	Ionosphere	351	32	126	35.90	Oryctognosy
19	landsat	6435	36	1333	20.71	Astronautics
20	letter	1600	32	100	6.25	Image
21	Lymphography	148	18	6	4.05	Healthcare
22	magic.gamma	19020	10	6688	35.16	Physical
23	mammography	11183	6	260	2.32	Healthcare
24	mnist	7603	100	700	9.21	Image
25	musk	3062	166	97	3.17	Chemistry
26	optdigits	5216	64	150	2.88	Image
27	PageBlocks	5393	10	510	9.46	Document
28	pendigits	6870	16	156	2.27	Image
29	Pima	768	8	268	34.90	Healthcare
30	satellite	6435	36	2036	31.64	Astronautics
31	satimage-2	5803	36	71	1.22	Astronautics
32	shuttle	49097	9	3511	7.15	Astronautics
33	skin	245057	3	50859	20.75	Image
34	smtp	95156	3	30	0.03	Web
35	SpamBase	4207	57	1679	39.91	Document
36	speech	3686	400	61	1.65	Linguistics
37	Stamps	340	9	31	9.12	Document
38	thyroid	3772	6	93	2.47	Healthcare
39	vertebral	240	6	30	12.50	Biology
40	vowels	1456	12	50	3.43	Linguistics
41	Waveform	3443	21	100	2.90	Physics
42	WBC	223	9	10	4.48	Healthcare
43	WDBC	367	30	10	2.72	Healthcare
44	Wilt	4819	5	257	5.33	Botany
45	wine	129	13	10	7.75	Chemistry
46	WPBC	198	33	47	23.74	Healthcare
47	yeast	1484	8	507	34.16	Biology
48	CIFAR10	5263	512	263	5.00	Image
49	FashionMNIST	6315	512	315	5.00	Image
50	MNIST-C	10000	512	500	5.00	Image
51	MVTec-AD	See Table B2.				Image
52	SVHN	5208	512	260	5.00	Image
53	Agnews	10000	768	500	5.00	NLP
54	Amazon	10000	768	500	5.00	NLP
55	Imdb	10000	768	500	5.00	NLP
56	Yelp	10000	768	500	5.00	NLP
57	20newsgroups	See Table B3.				NLP

---

[1] Ruff, Lukas, Robert A. Vandermeulen, Nico Görnitz, Alexander Binder, Emmanuel Müller, Klaus-Robert Müller, and Marius Kloft. "Deep semi-supervised anomaly detection. ICLR, 2019.

[2] Deecke, Lucas, et al. "Transfer-based semantic anomaly detection." ICML, 2021.

[3] Manolache, Andrei, Florin Brad, and Elena Burceanu. "DATE: Detecting Anomalies in Text via Self-Supervision of Transformers." NAACL, 2021.