Pith. sign in

REVIEW 4 major objections 6 minor 48 references

Tab2Visual: Overcoming Limited Data in Tabular Data Classification Using Deep Learning with Visual Representations

T0 review · 4 major / 6 minor · reviewed 2026-08-08 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Tab2Visual claims that converting tabular rows into proportional-width bar images lets a pretrained vision model beat all compared classifiers, including TabNet and TabPFN, on small datasets.

desk verdict A clean empirical study of a tabular-to-image method with useful ablations, but the headline claim of superiority on small datasets is not supported by the evidence as presented. read the letter →

arxiv 2502.07181 v1 pith:MWLHG6BW submitted 2025-02-11 cs.LG cs.CV

Ahmed Mamdouh , Moumen El-Melegy , Samia Ali , Ron Kikinis This is my paper · ORCID
classification cs.LGcs.CV
keywords tabulardatalimiteddeeplearningvisualrepresentationaugmentationtransferimageclassificationsmalldatasets
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Tab2Visual addresses a practical problem: deep learning models routinely lose to tree ensembles on small tabular datasets, yet many real-world datasets, especially in healthcare, have only a few hundred rows. The paper's proposal is to convert each tabular row into an image in which every feature is a colored vertical bar whose width is proportional to the normalized feature value, then fine-tune a pretrained EfficientNetV2 on those images. To stretch limited data, it adds elastic distortions and morphological operations designed to alter bar boundaries and widths while keeping the sample's class. Across the small datasets in the study, the augmented Tab2Visual variants achieved the best average F1-score and best average rank among all compared classifiers, including TabNet and TabPFN. If this holds, deep learning becomes usable on small tabular data without specialized tabular architectures.

What carries the argument

The central object is the Tab2Visual image: a canvas divided into rows and columns, one bar per feature, with bar width equal to the normalized feature value times a maximum width, and each feature assigned a distinct color. The mechanism is a two-step pipeline: first, tabular rows are normalized, one-hot encoded where needed, and drawn as bars; second, each image is optionally augmented by elastic distortion followed by random dilation, erosion, opening, or closing using randomly sized structuring elements. The augmented images then fine-tune a pretrained vision model whose feature weights are frozen except for the final classification layer. The bar layout lets a CNN read relative feature magnitudes as spatial extent.

What would settle it

A reader could apply Algorithm 2 to a set of Tab2Visual images, measure the bar widths back into feature vectors, and check whether the recovered values stay within the original class's observed feature ranges; if distorted images routinely turn near-zero features into visible bars or invert the ranking of two features, the augmentation is creating label noise rather than valid new samples.

Watch

Extended reading notes

Core claim

The central claim is that the visual format is the enabler: once tabular features are laid out as proportional-width bars in a compact image, a vision model pretrained on natural images can transfer its feature knowledge to the tabular task, and image-style augmentation can generate new training samples that stay close to the original distribution. The paper reports that on datasets with no more than about a thousand samples, Tab2Visual outperformed every compared method, with the version that augments each image four times posting the highest average F1-score and the best average rank, while CatBoost tied it on average AUC. On larger datasets, tree ensembles remained the best classifiers, so the claimed advantage is specifically for the limited-data regime.

Load-bearing premise

The load-bearing premise is that elastic distortion and morphological operations leave the class meaning of each bar image intact, so each augmented picture is a valid new sample of the same class.

Editorial extensions

If this is right

  • Small clinical and survey datasets with only a few hundred rows can be tackled with off-the-shelf vision models by rendering features as bar images, without building a new tabular architecture.
  • Transfer learning from natural-image models becomes a viable strategy for tabular data, at least when the input is a compact bar layout.
  • Image augmentation operations such as elastic distortion and morphological transformations can serve as tabular augmentation, expanding effective training size without collecting new records.
  • On larger tabular datasets, the method does not displace tree ensembles; its claimed advantage is specific to the limited-data regime.
  • Feature arrangement matters: one or two rows of bars outperforms taller layouts, so compact bar layouts should be the default configuration for this method.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The reported gains likely combine pretrained feature reuse, strong regularization, and augmentation, and the paper does not isolate these effects; a controlled comparison against the same backbone fine-tuned on the original tabular features would separate the visual encoding from the transfer-learning benefit.
  • The row-arrangement finding suggests the CNN is reading bars as a length-ranking code rather than as a spatial pattern; this predicts that deterministic orderings by feature correlation would behave differently from random orderings, which the paper leaves untested.
  • The method's dependence on pretrained natural-image features could be probed by comparing against a randomly initialized small CNN on larger augmented sets; the paper's from-scratch results on small data are much weaker, indicating transfer is doing real work.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper proposes Tab2Visual, a method that encodes each tabular sample as an image of colored bars whose widths are proportional to normalized feature values, then trains or fine-tunes an ImageNet-pretrained EfficientNetV2 (or EfficientViT) on these images, with optional image augmentations based on elastic distortion and morphological operations. The authors evaluate the method on ten UCI datasets, comparing it against classical classifiers, tree ensembles, TabNet, and TabPFN, using 5-fold cross-validation with macro-F1 and AUC. The central claim, stated in the abstract and conclusions, is that Tab2Visual outperforms all compared methods on datasets with at most 1000 samples.

Significance. If the claim were substantiated, Tab2Visual would be a practically useful contribution: it is conceptually simple, enables transfer learning and augmentation for tabular data, and the paper includes informative ablations on transfer learning, backbone choice, and feature arrangement. The algorithmic descriptions are clear and the benchmark covers ten datasets and a broad set of baselines. However, the evidence presented does not establish the claimed superiority. The winning Tab2Visual variant receives both heavy data augmentation and ImageNet pretraining that no baseline receives, the augmentation scale is selected after seeing the test results, and no statistical significance is reported. As a result, the paper's main conclusion is currently unsupported.

major comments (4)
  1. [Section 3.3 and Table 4 / Figure 6] The comparison is confounded. Tab2Visual uses ImageNet-pretrained EfficientNetV2-B0 and image augmentation, whereas none of the baselines receive transfer learning or any data augmentation. The reported advantage of Tab2Visual-A4 over CatBoost, MLP, and TabPFN on small datasets therefore cannot be attributed to the visual encoding; Table 5 shows that transfer learning alone yields a 7.5% average AUC gain on the small datasets. To support the central claim, the authors should compare against baselines that receive equivalent augmentation and transfer learning (e.g., SMOTE-augmented or otherwise pre-trained variants), or ablate Tab2Visual by removing augmentation and pretraining to isolate the contribution of the visual representation.
  2. [Section 4.2 and Table 4] The augmentation scale K is selected post hoc. The paper reports results for A0 through A4 on each small dataset and then highlights A4 in the abstract and conclusion, but K is not chosen by a validation set or nested cross-validation; it is the best-performing variant on the same test folds used for evaluation. This selection-on-test procedure inflates the reported performance and should be replaced with an independent validation-based selection of K or a pre-specified augmentation level.
  3. [Section 3.4, Table 4, and Figures 6-10] No confidence intervals, standard deviations, or significance tests are reported for any method. The margins that drive the conclusion are small (e.g., 74.6% vs 72.6% average F1 in Figure 6), and on average AUC CatBoost beats Tab2Visual-A4 (89.2% vs 88.7%). Without repeated runs or statistical testing, the claimed superiority of Tab2Visual on small datasets is not established.
  4. [Algorithm 2 and Section 2] The label-preserving property of the proposed augmentations is not validated. Elastic distortions and morphological dilation/erosion on bar-width images can alter feature values in ways that do not correspond to valid tabular samples; the claim that they generate 'synthetic samples that closely resemble the original data distribution' requires empirical support, for example by reconstructing feature values from augmented images and comparing their distributions. This issue is secondary to the confound above, but it matters because augmentation is the main source of the reported performance gain.
minor comments (6)
  1. [Algorithm 1 and Section 2] Algorithm 1 line 7 contains a typo: 'h ←= H/r' should be 'h ← H/r'. In Section 2, 'hight' should be 'height'.
  2. [Table 4] The column headers contain spacing artifacts ('HR T', 'SA T', 'JU', 'GL', 'PE') and the table is difficult to parse; consider a layout that separates F1 and AUC rows more clearly.
  3. [Section 3.4] The sentence 'data is first divided into non-overlapping training and testing partitions' is ambiguous in the context of 5-fold cross-validation; please specify how augmentation is applied within each fold's training partition.
  4. [Section 3.4] The paper reports AUC for multiclass datasets (Glass, Satellite) without specifying the averaging scheme; please clarify whether this is macro one-vs-rest AUC.
  5. [Figure 1] Figure 1 is presented without a description of the OpenML query or the date of access; please provide these details for reproducibility.
  6. [Section 4.2 and Figure 10] The claim of a 'consistent improvement' from augmentation is based on averages over six datasets without any measure of variance; showing per-dataset trajectories or error bars would be more informative.

Circularity Check

1 steps flagged · score 2.0 of 10

No construction-level circularity; the central empirical comparison is independent, with a minor self-citation and a test-set-selected augmentation variant noted as mild concerns.

  1. other [Section 1, contribution paragraph]
    "This work significantly extends the preliminary, exploratory ideas drafted in our prior work [29], which focused on a specific clinical context."

    The paper's core visual-encoding idea is inherited from the authors' own prior work [29], so the 'novel approach' claim leans on an internal source. However, [29] is not used as evidence for the current benchmark outcomes: the experiments, baselines, and ablations are new and externally checkable, and no uniqueness theorem or derivation is imported. Thus the self-citation is minor and not load-bearing; it is noted for scoring rather than as a construction-level circularity.

full rationale

The paper contains no derivation chain whose output is equivalent to its input by construction. Tab2Visual's pipeline—normalization, bar-width image encoding, elastic/morphological augmentation, and pretrained-backbone fine-tuning—is an empirical design; every claimed advantage is backed by held-out fold measurements reported in Tables 4–7. The only internal-source dependency is the visual-encoding concept from the authors' prior work [29], which is explicitly acknowledged; that citation is not load-bearing because the current study's comparisons, ablations, and augmentations are newly executed and can be checked against the reported tables. Two methodological concerns exist but are not circularity: the augmentation level K is reported for A0–A4 and the best variant (A4) is highlighted after the fact, and the conclusion 'Tab2Visual has outperformed' omits that it is specifically the A4 variant; also, baselines receive no augmentation or transfer learning, and no significance tests are reported. These issues affect the strength of the empirical claim, but they do not make the result equivalent to its inputs: A4's margins over CatBoost, MLP, and TabPFN are contingent measurements, not consequences of the selection rule. Score 2 reflects the minor self-citation; no higher circularity score is warranted.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

The central claim rests on a chain of empirical choices: the image encoding, the choice of ImageNet transfer, the specific augmentation operations, and the post-hoc selection of augmentation level. None of these are derived from first principles or validated with significance tests. The free parameters are mostly hand-set ranges and Optuna-tuned hyperparameters. No new entities (particles, mediators, forces) are introduced.

free parameters (6)
  • Augmentation scale K = 4 (A4)
    Chosen as the best level after inspecting average F1/AUC across A0-A4 on small datasets (Section 4.2, Figure 10).
  • Elastic distortion parameters alpha, sigma = alpha in [40,60], sigma in [3,5]
    Hand-set ranges in Section 3.3; not optimized per dataset and not reported as tuned.
  • Morphological probabilities Pd, Pe and SE sizes = Pd, Pe in [0.6,0.8]; SE up to (2,5)
    Hand-set ranges in Section 3.3.
  • Number of rows r = 1
    Fixed to 1 for all experiments; Section 4.5 shows r=1 or 2 is best on the Satellite dataset, so the choice is empirically motivated but not per-dataset tuned.
  • Image dimensions H, W = 224 x 224
    Set to match EfficientNetV2 input requirements (Section 3.3).
  • Fine-tuning hyperparameters (learning rate, batch size, weight decay, dropout, hidden units) = lr=0.0005, batch=32, wd=0.005, dropout=0.5, 1 hidden layer of 128 units (US Election best config)
    Tuned with Optuna on cross-validation folds (Table 3), so these are fitted to the data.
assumptions (4)
  • domain assumption The bar-width image encoding preserves the discriminative information of the tabular features for a CNN or ViT to learn.
    Section 2, Feature Encoding; no proof that spatial arrangement and color coding add useful inductive bias; the encoding is a hand-designed choice.
  • domain assumption ImageNet-pretrained weights transfer to bar-chart-like images.
    Section 3.3 uses EfficientNetV2-B0 pretrained on ImageNet-1K and freezes all but the final layer; the domain gap to synthetic bar images is untested.
  • domain assumption Elastic distortion and morphological operations preserve the class label.
    Algorithm 2 and Section 2; the paper asserts the augmented images 'closely resemble the original data distribution' but provides no label-preservation check.
  • domain assumption The six datasets with fewer than 1000 rows are representative of 'limited tabular data'.
    Section 3.1; the selection is convenient, and the OpenML claim (71% <10K) does not establish representation of the small-data regime.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Tab2Visual: Overcoming Limited Data in Tabular Data Classification Using Deep Learning with Visual Representations." pith.science (2026). https://pith.science/paper/MWLHG6BW

@misc{pith2026250207181,
  author       = {Pith},
  title        = {Pith review of: Tab2Visual: Overcoming Limited Data in Tabular Data Classification Using Deep Learning with Visual Representations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MWLHG6BW}},
  note         = {Machine review of arXiv:2502.07181}
}
read the original abstract

This research addresses the challenge of limited data in tabular data classification, particularly prevalent in domains with constraints like healthcare. We propose Tab2Visual, a novel approach that transforms heterogeneous tabular data into visual representations, enabling the application of powerful deep learning models. Tab2Visual effectively addresses data scarcity by incorporating novel image augmentation techniques and facilitating transfer learning. We extensively evaluate the proposed approach on diverse tabular datasets, comparing its performance against a wide range of machine learning algorithms, including classical methods, tree-based ensembles, and state-of-the-art deep learning models specifically designed for tabular data. We also perform an in-depth analysis of factors influencing Tab2Visual's performance. Our experimental results demonstrate that Tab2Visual outperforms other methods in classification problems with limited tabular data.

Figures

Figures reproduced from arXiv: 2502.07181 by the authors.

Figure 1
Figure 1. Distribution of tabular dataset sizes on the OpenML platform ( [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. The Tab2Visual approach: Tabular data undergoes normalization and is subse [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Visual representation of a sample from a 9-feature tabular dataset. The bar [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Tab2Visual representations of a sample with 40 features arranged in different [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]
Figure 5
Figure 5. Figure 5: Example of data augmentation applied to Tab2Visual image representations of [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]
Figure 6
Figure 6. Figure 6: Average F1-score and AUC of different classification algorithms on smaller [PITH_FULL_IMAGE:figures/full_fig_p023_6.png]
Figure 7
Figure 7. Figure 7: Average F1-score and AUC of different classification algorithms on larger [PITH_FULL_IMAGE:figures/full_fig_p023_7.png]
Figure 8
Figure 8. Figure 8: Average rank of classification algorithms based on F1-score and AUC across [PITH_FULL_IMAGE:figures/full_fig_p024_8.png]
Figure 9
Figure 9. Figure 9: Average rank of classification algorithms based on F1-score and AUC across [PITH_FULL_IMAGE:figures/full_fig_p024_9.png]
Figure 10
Figure 10. Figure 10: Impact of augmentation levels (from A0 to A4) on Tab2Visual’s average per￾formance on smaller datasets. Performance is measured in terms of F1-score and AUC. Higher values indicate better performance. instance, AUC accuracy increases from 86.6% at A0 to 88.7% at A4, r…
Figure 11
Figure 11. Figure 11: Impact of feature arrangement on Tab2Visual’s performance: F1-score and [PITH_FULL_IMAGE:figures/full_fig_p028_11.png]
Figure 12
Figure 12. Figure 12: Computational cost analysis: Heatmap of average 5-fold training times (in [PITH_FULL_IMAGE:figures/full_fig_p029_12.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

48 extracted references · 39 canonical work pages

  1. [29]

    El-Melegy, A

    M. El-Melegy, A. Mamdouh, S. Ali, M. Badawy, M. A. El-Ghar, N. S. Alghamdi, A. El-Baz, Prostate cancer diagnosis via visual representation of tabular data and deep transfer learning, Bioengineering 11 (7) (2024) 635

    work page 2024

  2. [1]

    Krizhevsky, I

    A. Krizhevsky, I. Sutskever, G. E. Hinton, Imagenet classification with deep convolutional neural networks, in: Advances in Neural Information Processing Systems, Vol. 25, 2012

    work page 2012

  3. [2]

    K. He, X. Zhang, S. Ren, J. Sun, Deep residual learning for image recognition, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016, pp. 770–778

    2016

  4. [3]

    Dosovitskiy, L

    A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Unterthiner, et al., An image is worth 16x16 words: Transformers for image recognition at scale, arXiv preprint arXiv:2010.11929 (2020)

    arXiv 2020

  5. [4]

    Carion, F

    N. Carion, F. Massa, G. Synnaeve, N. Usunier, A. Kirillov, S. Zagoruyko, End-to-end object detection with transformers, in: European Conference on Computer Vision, Springer, 2020, pp. 213–229

    work page 2020

  6. [5]

    Silver, J

    D. Silver, J. Schrittwieser, K. Simonyan, I. Antonoglou, A. Huang, A. Guez, et al., Mastering the game of Go without human knowledge, Nature 550 (7676) (2017) 354–359

    work page 2017

  7. [6]

    Grinsztajn, E

    L. Grinsztajn, E. Oyallon, G. Varoquaux, Why do tree-based models still outperform deep learning on typical tabular data?, in: Thirty-sixth Conference on Neural Information Processing Systems Datasets and Benchmarks Track, 2022

    work page 2022

  8. [7]

    Borisov, T

    V. Borisov, T. Leemann, K. Seßler, J. Haug, M. Pawelczyk, G. Kasneci, Deep neural networks and tabular data: A survey, IEEE Transactions on Neural Networks and Learning Systems 35 (6) (2024) 7499–7519. doi:10.1109/TNNLS.2022.3229161

    work page arXiv 2024

Show all 48 references
  1. [8]

    Van Breugel, M

    B. Van Breugel, M. Van Der Schaar, Position: Why tabular foundation models should be a research priority, in: Proceedings of the 41st International Conference on Machine Learning, ICML’24, 2024

    2024

  2. [9]

    Popov, S

    S. Popov, S. Morozov, A. Babenko, Neural oblivious decision ensembles for deep learning on tabular data, in: International Conference on Learning Representations, 2020

    2020

  3. [10]

    S. ¨O. Arik, T. Pfister, TabNet: Attentive interpretable tabular learning, in: Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 35, 2021, pp. 6679–6687

    2021

  4. [11]

    Huang, A

    X. Huang, A. Khetan, M. Cvitkovic, Z. Karnin, TabTransformer: Tabular data modeling using contextual embeddings, arXiv preprint arXiv:2012.06678 (2020)

    2020 arXiv

  5. [12]

    Hollmann, S

    N. Hollmann, S. M¨ uller, K. Eggensperger, F. Hutter, TabPFN: A transformer that solves small tabular classification problems in a second, in: Proceedings of The Eleventh International Conference on Learning Representations, 2023

    2023

  6. [13]

    Hollmann, S

    N. Hollmann, S. M¨ uller, L. Purucker, A. Krishnakumar, M. K¨ orfer, S. Hoo, R. Schirrmeister, F. Hut- ter, Accurate predictions on small data with a tabular foundation model, Nature 637 (2025) 319–326. doi:10.1038/s41586-024-08328-6 . 32

    2025 doi

  7. [14]

    Shavitt, E

    I. Shavitt, E. Segal, Regularization learning networks: deep learning for tabular datasets, in: Pro- ceedings of the 32nd International Conference on Neural Information Processing Systems, NIPS’18, Curran Associates Inc., Red Hook, NY, USA, 2018, p. 1386–1396

    2018

  8. [15]

    Kadra, M

    A. Kadra, M. Lindauer, F. Hutter, J. Grabocka, Well-tuned simple nets excel on tabular datasets, in: Proceedings of the 35th International Conference on Neural Information Processing Systems, NIPS ’21, Curran Associates Inc., Red Hook, NY, USA, 2021

    2021

  9. [16]

    Sharma, E

    A. Sharma, E. Vans, D. Shigemizu, K. A. Boroevich, T. Tsunoda, Deepinsight: A methodology to transform a non-image data to an image for convolution neural network architecture, Scientific reports 9 (1) (2019) 11399

    2019

  10. [17]

    van der Maaten, G

    L. van der Maaten, G. Hinton, Visualizing data using t-SNE, Journal of Machine Learning Research 9 (86) (2008) 2579–2605

    2008

  11. [18]

    Bazgir, R

    O. Bazgir, R. Zhang, S. R. Dhruba, R. Rahman, S. Ghosh, R. Pal, Representation of features as images with neighborhood dependencies for compatibility with convolutional neural networks, Nature Communications 11 (09 2020). doi:10.1038/s41467-020-18197-y

    2020 doi

  12. [19]

    B. Sun, L. Yang, W. Zhang, M. Lin, P. Dong, C. Young, J. Dong, SuperTML: Two-dimensional word embedding for the precognition on structured tabular data, in: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPR W), 2019, pp. 2973–2981. doi: 10.110...

    2019

  13. [20]

    Buturovi´ c, D

    L. Buturovi´ c, D. Miljkovi´ c, A novel method for classification of tabular data using convolutional neural networks, BioRxiv (2020) 2020–05

    2020

  14. [21]

    Nguyen, H

    T. Nguyen, H. Tran, Image generator for tabular data (IGTD), Pattern Recognition Letters 145 (2021) 71–78

    2021

  15. [22]

    Shwartz-Ziv, A

    R. Shwartz-Ziv, A. Armon, Tabular data: Deep learning is not all you need, Information Fusion 81 (2022) 84–90. doi:https://doi.org/10.1016/j.inffus.2021.11.011

    2022 doi

  16. [23]

    M. Xu, S. Yoon, A. Fuentes, D. S. Park, A comprehensive survey of image augmentation techniques for deep learning, Pattern Recognition 137 (2023) 109347. doi:https://doi.org/10.1016/j.patcog. 2023.109347

    2023

  17. [24]

    C. Tan, F. Sun, T. Kong, W. Zhang, C. Yang, C. Liu, A survey on deep transfer learning, in: V. K ˚ urkov´ a, Y. Manolopoulos, B. Hammer, L. Iliadis, I. Maglogiannis (Eds.), Artificial Neural Net- works and Machine Learning – ICANN 2018, Springer International Publishing, Cham,...

    2018

  18. [25]

    N. V. Chawla, K. W. Bowyer, L. O. Hall, W. P. Kegelmeyer, SMOTE: synthetic minority over- sampling technique, Journal of Artificial Intelligence Research 16 (1) (2002) 321–357

    2002

  19. [26]

    Darabi, Y

    S. Darabi, Y. Elor, Synthesising multi-modal minority samples for tabular data, ArXiv abs/2105.08204 (2021)

    2021 arXiv

  20. [27]

    Mamdouh, M

    A. Mamdouh, M. T. El-Melegy, S. A. Ali, A. S. El-Baz, Prediction of the gleason group of prostate cancer from clinical biomarkers: Machine and deep learning from tabular data, in: 2022 International Joint Conference on Neural Networks (IJCNN), IEEE, 2022, pp. 1–8

    2022

  21. [28]

    M. T. El-Melegy, A. Mamdouh, S. A. Ali, A. S. El-Baz, Prostate cancer diagnosis from structured clinical biomarkers with deep learning, in: 2022 International Conference on Digital Image Comput- ing: Techniques and Applications (DICTA), IEEE, 2022, pp. 1–8. 33

    2022

  22. [30]

    Paivio, K

    A. Paivio, K. Csapo, Picture superiority in free recall: Imagery or dual coding?, Cognitive Psychology 5 (2) (1973) 176–206. doi:https://doi.org/10.1016/0010-0285(73)90032-7

    1973 doi

  23. [31]

    M. Z. Mintzer, J. G. Snodgrass, The picture superiority effect: Support for the distinctiveness model, The American Journal of Psychology 112 (1) (1999) 113–146

    1999

  24. [32]

    C. L. Grady, A. R. McIntosh, M. N. Rajah, F. I. M. Craik, Neural correlates of the episodic encoding of pictures and words, Proceedings of the National Academy of Sciences 95 (5) (1998) 2703–2708. doi:10.1073/pnas.95.5.2703

    1998 doi

  25. [33]

    M. Tan, Q. Le, EfficientNet: Rethinking model scaling for convolutional neural networks, in: Inter- national conference on machine learning, PMLR, 2019, pp. 6105–6114

    2019

  26. [34]

    X. Liu, H. Peng, N. Zheng, Y. Yang, H. Hu, Y. Yuan, EfficientViT: Memory efficient vision trans- former with cascaded group attention, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 14420–14430

    2023

  27. [35]

    Buslaev, V

    A. Buslaev, V. I. Iglovikov, E. Khvedchenya, A. Parinov, M. Druzhinin, A. A. Kalinin, Albumenta- tions: Fast and flexible image augmentations, Information 11 (2) (2020)

    2020

  28. [36]

    T. K. Ho, Random decision forests, in: Proceedings of 3rd international conference on document analysis and recognition, Vol. 1, IEEE, 1995, pp. 278–282

    1995

  29. [37]

    Geurts, D

    P. Geurts, D. Ernst, L. Wehenkel, Extremely randomized trees, Machine learning 63 (2006) 3–42

    2006

  30. [38]

    T. Chen, C. Guestrin, XGBoost: A scalable tree boosting system, in: Proceedings of the 22nd ACM sigkdd international conference on knowledge discovery and data mining, 2016, pp. 785–794

    2016

  31. [39]

    G. Ke, Q. Meng, T. Finley, T. Wang, W. Chen, W. Ma, Q. Ye, T.-Y. Liu, LightGBM: A highly efficient gradient boosting decision tree, Advances in neural information processing systems 30 (2017)

    2017

  32. [40]

    A. V. Dorogush, V. Ershov, A. Gulin, CatBoost: Gradient boosting with categorical features support, arXiv preprint arXiv:1810.11363 (2018)

    2018 arXiv

  33. [41]

    M. Tan, Q. Le, EfficientNetV2: Smaller models and faster training, in: International conference on machine learning, PMLR, 2021, pp. 10096–10106

    2021

  34. [42]

    L. Deng, H. Suo, D. Li, et al., Deepfake video detection based on EfficientNet-V2 network, Compu- tational Intelligence and Neuroscience 2022 (2022)

    2022

  35. [43]

    Banerjee, M

    S. Banerjee, M. K. H. Monir, CEIMVEN: An approach of cutting edge implementation of modi- fied versions of efficientnet (v1-v2) architecture for breast cancer detection and classification from ultrasound images, arXiv preprint arXiv:2308.13356 (2023)

    2023 arXiv

  36. [44]

    Wightman, Pytorch image models, https://github.com/huggingface/pytorch-image-models (2019)

    R. Wightman, Pytorch image models, https://github.com/huggingface/pytorch-image-models (2019). doi:10.5281/zenodo.4414861

    2019 doi

  37. [45]

    Akiba, S

    T. Akiba, S. Sano, T. Yanase, T. Ohta, M. Koyama, Optuna: A next-generation hyperparameter optimization framework, in: Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining, 2019, pp. 2623–2631. 34

    2019

  38. [46]

    Pedregosa, G

    F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Pret- tenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, E. Duchesnay, Scikit-learn: Machine learning in Python, Journal of Machine Learning R...

    2011

  39. [47]

    C. Hou, J. Zhang, T. Zhou, When to learn what: Model-adaptive data augmentation curriculum, in: 2023 IEEE/CVF International Conference on Computer Vision (ICCV), 2023, pp. 1717–1728. doi:10.1109/ICCV51070.2023.00165

    2023

  40. [48]

    Cheung, D.-Y

    T.-H. Cheung, D.-Y. Yeung, AdaAug: Learning class- and instance-adaptive data augmentation policies, in: International Conference on Learning Representations, 2022. 35

    2022

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.