Sample stimuli

How to use

from brainscore_vision import load_benchmark
benchmark = load_benchmark("Marques2020_Schiller1976-sf_selective")
score = benchmark(my_model)

Benchmark API

Code examples

Model scores

Score Legend Min Alignment Max Alignment

Rank	Model	Score
1	imagenet_l2_10_0	1.0
2	resnext101_32x8d_wsl	1.000
3	BiT-M-R101x1	1.0
4	resnet-50-robust	1.0
5	mobilenet_v2_0.75_160-tf	1.000
6	resnet34_ecoset_full	1.0
7	barlow-twins-resnet50	1.000
8	yudixie_resnet50_distance_translation_rotation_0_240908	.999
9	fulltest_microblockvf_nobottleneck_freeat_nopretrain_eps2_m4	.999
10	alexnet_robust_correct	.999
11	resnet50-vitoimagevidnet8	.999
12	mobilenet_v2_1.0_96	.999
13	resnet50_robust_l2_eps1	.998
14	resnet50_robust_l2_eps3	.999
15	resnet50_l2_3_robust	.996
16	grcnn_v2_text_noise	.999
17	resnet-18-LC_w_sh_10_iter_m	.996
18	resnet50_linf_4_robust	.995
19	blt_vs	.996
20	alexnet_linf_8_robust	.996
21	yudixie_resnet50_category_class_0_240908	.996
22	AdvProp_efficientnet-b8	.998
23	yudixie_resnet50_rotation_reg_0_240908	.997
24	resnet50-moclr8deg	.997
25	resnet50-SIN	.995
26	yudixie_resnet18_cat_obj_class_all_latents_0_240719	.993
27	resnet50_linf_8_robust	.994
28	effnetb1_cutmixpatch_SAM_robust32_avge6e8e9e10_manylayers_324x288	.992
29	vit_relpos_base_patch32_plus_rpn_256:sw_in1k	.994
30	effnetb1_cutmixrespatch_SAM_robust32_e2_manylayers_324x288	.993
31	effnetb1_cutmixpatch_augmix_robust32_avge4e7_manylayers_324x288	.993
32	effnetb1_cutmixrespatch_SAM_robust32_e10_manylayers_324x288	.991
33	resnet-18-LC_1st_conv_m	.992
34	effnetb1_cutmixpatch_augmix_robust32_e5_324x288	.991
35	antialias-resnet152	.990
36	CORnetZ_CIFAR10	.991
37	alexnet_l2_3_robust	.990
38	resnet50_ecoset_full	.990
39	resnet101	.989
40	resnet50_finetune_cutmix_e3_robust_linf8255_e0_247x234	.989
41	BiT-S-R152x2	.990
42	resnet50_FractalDB	.987
43	alexnet_training_seed_06	.989
44	vgg-16-keras	.987
45	resnet50-simclr-vissl	.985
46	texture_shape_resnet50_trained_on_SIN	.984
47	vgg-19-keras	.985
48	resnet50_finetune_cutmix_AVGe2e3_robust_linf8255_e0_247x234	.987
49	resnet18_imagenet_full	.984
50	resnet101_ecoset_full	.984
51	alexnet_training_seed_08	.983
52	imagenet_l2_3_0	.984
53	AdvProp_efficientnet-b4	.985
54	vgg_16	.981
55	mobilenet_v2_1.0_128-tf	.980
56	resnet-50_untrained	.979
57	yudixie_resnet50_object_class_0_240908	.978
58	alexnet_training_seed_04	.975
59	yudixie_resnet50_distance_rotation_0_240908	.976
60	yudixie_resnet18_category_class_0_240719	.972
61	cornet_s	.972
62	resnet50_imagenet_10_seed-0	.969
63	voneresnet-50-non_stochastic	.970
64	inception_v1	.969
65	yudixie_resnet18_distance_reg_0_240719	.968
66	BiT-M-R50x3	.969
67	yudixie_resnet50_distance_translation_0_240908	.967
68	resnet-152_v2_pytorch	.966
69	yudixie_resnet18_translation_rotation_0_240719	.965
70	resnet34_imagenet_full	.967
71	yudixie_resnet18_distance_translation_0_240719	.966
72	resnet_101_v1	.964
73	BiT-S-R50x1	.964
74	voneresnet-50-robust	.963
75	Res2Net50_26w_4s	.960
76	alexnet_training_seed_09	.958
77	voneresnet-50	.959
78	r3m_resnet18	.955
79	grcnn_robust_v1	.954
80	resnet-18-LC_w_sh_1_iter_conv_init_m	.951
81	VOneCORnet-S	.954
82	resnet50-cifar	.951
83	yudixie_resnet50_cat_obj_class_all_latents_0_240908	.950
84	mobilevit_small	.947
85	yudixie_resnet50_translation_rotation_0_240908	.947
86	yudixie_resnet50_translation_reg_0_240908	.945
87	resnet50-vicregl0p9	.945
88	alexnet_training_seed_03	.944
89	resnet18_ecoset_full	.944
90	inception_v2-tf	.944
91	yudixie_resnet18_random_0_240719	.938
92	vonegrcnn_52e_full	.939
93	resnet-18-LC_w_sh_100_iter_conv_init_m	.936
94	yudixie_resnet18_object_class_0_240719	.937
95	resnet101_imagenet_full	.936
96	omnivore_swinB	.934
97	yudixie_resnet50_distance_reg_0_240908	.929
98	mobilenet_v2_1.0_160-tf	.928
99	antialiased-r50	.927
100	resnet50_imagenet_100_seed-0	.928
101	mobilenet_v2_0.5_192-tf	.926
102	resnet-18-LC_w_sh_100_iter_m	.924
103	briaai_rmbg_1_4	.925
104	grcnn_109	.924
105	CLIP-RN50	.925
106	resnet-18_untrained	.922
107	vggface	.921
108	resnet-18-LC_m	.921
109	yudixie_resnet18_distance_rotation_0_240719	.919
110	yudixie_resnet18_translation_reg_0_240719	.919
111	0.5x_resnet-18	.918
112	focalnet_tiny_lrf_in1k	.917
113	vonegrcnn_47e	.916
114	inception_v1-tf	.916
115	alexnet_training_seed_10	.914
116	vision_transformer_vit_large_patch16_224	.915
117	alexnet_random_linf8_perturb	.912
118	yudixie_resnet18_distance_translation_rotation_0_240719	.912
119	alexnet_training_seed_02	.907
120	resnet50-SIN_IN	.907
121	vonealexnet_gaussian_noise_std4_fixed	.905
122	0.5x_resnet-18_LC_w_sh_100_iter	.904
123	resnet-50x2_untrained	.902
124	yudixie_resnet50_random_0_240908	.901
125	mobilenet_v2_0.75_96	.900
126	grcnn	.898
127	resnet50-vicregl0p75	.898
128	xception	.896
129	alexnet_training_seed_01	.894
130	alexnet_early_checkpoint	.893
131	artResNet18_1	.893
132	resnet_101_v2	.891
133	resnet152_ecoset_full	.890
134	alexnet	.888
135	resnet-50_v1-tf	.883
136	convnext_small_imagenet_1_seed-0	.883
137	densenet-169	.882
138	resnet-18-LC_conv_init_m	.879
139	ViT_L_32_imagenet1k	.881
140	vonegrcnn_47e	.879
141	resnet-18-LC_w_sh_10_iter_conv_init_m	.876
142	tf_efficientnetv2_s_in21ft1k_robust_linf12255_400x400	.878
143	r3m_resnet34	.877
144	resnet50_imagenet_full	.876
145	konkle_alexnetgn_ipcl_ref12_supervised_ipcl_aug	.873
146	texture_shape_alexnet_trained_on_SIN	.870
147	yudixie_resnet18_imagenet1kpret_0_240719	.870
148	resnet50_tutorial	.867
149	yudixie_resnet50_imagenet1kpret_0_240908	.866
150	0.5x_resnet-18_LC	.865
151	resnet50_byol	.862
152	resnet_50_v1	.863
153	resnet50-sup	.861
154	alexnet_random_l2_3_perturb	.858
155	resnet_50_v2	.858
156	alexnet_training_seed_07	.858
157	resnet-18_test_m	.857
158	mobilenet_v2_1_0_192	.857
159	yudixie_resnet18_rotation_reg_0_240719	.857
160	vonegrcnn_62e_nobn	.856
161	r3m_resnet50_nocrop	.852
162	convnext_femto_ols:d1_in1k	.852
163	resnet-18-LC_w_sh_1_iter_m	.845
164	ReAlnet10	.845
165	ReAlnet10_cornet	.844
166	CLIP_resnet50	.841
167	BiT-M-R50x1	.841
168	alexnet	.840
169	CORnet-Z	.840
170	hmax	.839
171	mobilenet_v2_0.75_128	.840
172	mobilenet_v2_0.5_128	.840
173	0.5x_resnet-18_LC_w_sh_10_iter	.841
174	alexnet	.840
175	0.5x_resnet-18_LC_w_sh_1_iter	.840
176	alexnet-baseline	.840
177	alexnet	.839
178	alexnet	.841
179	resnet-18-LC_d_w_sh_1x1_conv_init_m	.838
180	alexnet	.840
181	CORnetZ_CIFAR10_bs32_20_04	.840
182	CORnet-S	.839
183	alexnet_reduced_aliasing_early_checkpoint	.839
184	alexnet	.840
185	alexnet	.839
186	ReAlnet06_cornet	.839
187	ReAlnet04_cornet	.841
188	ReAlnet03	.839
189	ReAlnet07_cornet	.840
190	ReAlnet07	.840
191	ReAlnet06	.840
192	ReAlnet04	.841
193	ReAlnet08_cornet	.838
194	ReAlnet01	.840
195	AlexNet_SIN	.839
196	ReAlnet05	.839
197	ReAlnet09_cornet	.839
198	ReAlnet03_cornet	.840
199	ReAlnet02	.840
200	ReAlnet02_cornet	.839
201	ReAlnet01_cornet	.839
202	ReAlnet_incorrect	.839
203	ReAlnet08	.840
204	ReAlnet05_cornet	.840
205	ReAlnet09	.838
206	resnet50_pretrained_with_retinal_waves	.838
207	CLIP_resnet50_float32	.837
208	mobilenet_v2_1_4_224	.835
209	mobilenet_v2_1_0_160	.836
210	densenet-169-keras	.831
211	resnet50-SIN_IN_IN	.827
212	resnet50-vicreg	.824
213	BiT-M-R152x2	.825
214	AdvProp_efficientnet-b0	.825
215	mobilenet_v1_0.25_128	.821
216	resnet50-barlow	.820
217	cvt_cvt-13-224-in1k_4_LucyV4	.816
218	vit_tiny_r_s16_p8_384:augreg_in21k_ft_in1k	.814
219	efficientnet_b2_imagenet_full	.814
220	RN50	.813
221	vgg_19	.805
222	efficientnet-b7	.804
223	nasnet_large-tf	.804
224	BiT-M-R152x4	.802
225	mobilenet_v2_0_75_192	.800
226	BiT-S-R101x1	.800
227	mobilenet_v2_0.75_224-tf	.799
228	resnet50-VITO-8deg-cc	.799
229	AdvProp_efficientnet-b7	.800
230	AdvProp_efficientnet-b6	.796
231	mobilenet_v2_1.0_224-tf	.791
232	SWSL_resnext101_32x8d	.782
233	my-model	.782
234	mobilenet_v2_1_0_224	.773
235	resnet50_simclr	.774
236	BiT-M-R101x3	.773
237	resnet50	.767
238	effnetb1_272x240	.768
239	inception_v4-tf	.769
240	resnet50	.767
241	resnet-152_v1-tf	.763
242	SWSL_resnet50	.759
243	densenet-201	.761
244	vit_base_patch16_clip_224:openai_ft_in12k_in1k	.756
245	densenet-121	.754
246	inception_v3	.749
247	omnivore_swinS	.748
248	resnet50_random_linf8_perturb	.743
249	resnet152_imagenet_full	.740
250	convnext_small_imagenet_10_seed-0	.735
251	r3m_resnet50	.733
252	convnext_small_imagenet_full_seed-0	.732
253	v1-pyr-nodown	.731
254	resnet18_imagenet21kP	.728
255	convnext_large_imagenet_full_seed-0	.726
256	resnet50_random_l2_perturb	.721
257	pnasnet_large-tf	.722
258	mobilenet_v2_0.75_192-tf	.720
259	nasnet_mobile-tf	.720
260	konkle_alexnetgn_ipcl_ref01_primary_model	.720
261	mobilenet_v2_0_5_224	.718
262	alexnet_training_seed_05	.710
263	CLIP_ViT-B_32	.708
264	resnet_SIN_IN_FT_IN	.694
265	mobilenet_v2_0.5_224-tf	.694
266	nasnet_mobile	.694
267	efficientnet_b0	.691
268	deit_base_imagenet_full_seed-0	.692
269	deit_small_imagenet_full_seed-0	.688
270	inception_v3-tf	.688
271	BiT-S-R50x3	.687
272	yolos_tiny	.685
273	cvt_cvt-13-384-in22k_finetuned-in1k_4_LucyV4	.683
274	convnext_tiny_imagenet_full_seed-0	.679
275	efficientnet_b1_imagenet_full	.679
276	my-model	.672
277	vit_base_patch16_clip_224:openai_ft_in1k	.671
278	effnetb1_cutmix_augmix_sam_e1_5avg_424x377	.664
279	BiT-S-R152x4	.661
280	mobilenet_v2_1.4_224-tf	.660
281	mobilenet_v2_1.0_192-tf	.653
282	convnext_large:fb_in22k_ft_in1k	.647
283	densenet-201-keras	.643
284	densenet-121-keras	.641
285	convnext_xlarge:fb_in22k_ft_in1k	.640
286	mobilenet_v2_1-4_224_pytorch	.639
287	cvt_cvt-21-224-in1k_4_LucyV4	.638
288	resnet50_moco_v2	.637
289	mobilenet_v2_1.3_224-tf	.636
290	mobilenet_v2_0_5_192	.636
291	FrankRobWobv0	.623
292	mobilenet_v2_0.5_160	.623
293	resnet-101_v2-tf	.620
294	deit_small_imagenet_10_seed-0	.618
295	mobilenet_v2_1_0_128	.617
296	convnext_tiny_sup	.618
297	antialiased-rnext101_32x8d	.613
298	convnext_xxlarge:clip_laion2b_soup_ft_in1k	.608
299	vit_relpos_base_patch16_clsgap_224:sw_in1k	.605
300	convnext_tiny:in12k_ft_in1k	.603
301	efficientnet_b0_imagenet_full	.600
302	effnetv2m_custom384	.599
303	resnet-101_v1-tf	.584
304	mobilenet_v2_1_3_224	.585
305	mobilenet_v2_0_75_160	.583
306	deit_small_imagenet_100_seed-0	.580
307	resnet-50_v2-tf	.579
308	BiT-S-R101x3	.575
309	cvt_cvt-21-384-in1k_4_LucyV4	.575
310	cvt_cvt-13-384-in1k_4_LucyV4	.571
311	vit_huge_patch14_clip_224:laion2b_ft_in12k_in1k	.557
312	resnet-152_v2-tf	.557
313	AT_efficientnet-b2	.550
314	AdvProp_efficientnet-b2	.549
315	convnext_base_imagenet_full_seed-0	.543
316	deit_large_imagenet_full_seed-0	.543
317	inception_resnet_v2	.541
318	vit_huge_patch14_clip_336:laion2b_ft_in12k_in1k	.540
319	convnext_large_mlp:clip_laion2b_augreg_ft_in1k_384	.535
320	shufflenet_v2_x1_0	.534
321	cvt_cvt-w24-384-in22k_finetuned-in1k_4	.532
322	vit_large_patch14_clip_336:openai_ft_in12k_in1k	.527
323	ViT-B/32	.520
324	custom_model_cv_18_dagger_408	.513
325	mobilenet_v2_0.35_224	.498
326	vit_large_patch14_clip_336:laion2b_ft_in1k	.467
327	vit_large_patch14_clip_224:laion2b_ft_in1k	.465
328	omnivore_swinT	.465
329	cv_18_dagger_408_pretrained	.462
330	mobilenet_v2_0_75_224	.456
331	convnext_small_imagenet_100_seed-0	.457
332	resnet50_imagenet_1_seed-0	.443
333	vit_large_patch14_clip_224:laion2b_ft_in12k_in1k	.437
334	resnet-18-LC_untrained	.414
335	convnext_base:clip_laiona_augreg_ft_in1k_384	.413
336	vit_large_patch14_clip_224:openai_ft_in12k_in1k	.409
337	vit_large_patch14_clip_224:openai_ft_in1k	.404
338	swin_small_patch4_window7_224:ms_in22k_ft_in1k	.395
339	cvt_cvt-21-384-in22k_finetuned-in1k_4_LucyV4	.346
340	resnet50-meshes-lt-100-original-pretrained	.161
341	resnet50-meshes-lt-100-original-scratch	.063
342	resnet_152_v1	X
343	yudixie_resnet50_imagenet1kpret_0_240312	X
344	alexnet_ks_torevert	X
345	tv_efficientnet-b1	X
346	regnet_y_400mf	X
347	resnet_152_v2	X
348	resnext101_32x32d_wsl	X
349	resnext101_32x16d_wsl	X
350	resnext101_32x48d_wsl	X
351	pnasnet_large	X
352	resnet50_ImageNet	X
353	deit_small_imagenet_1_seed-0	X
354	nasnet_large	X
355	mobilenet_v2_0.35_160	X
356	vonegrcnn_62e_nobn	X
357	inception_v4	X
358	resnet18-supervised	X
359	inception_v3_pytorch	X
360	pnasnet_large_pytorch	X
361	densenet_201_pytorch	X
362	vonegrcnn_52e_full	X
363	fitvid_trained_on_physion	X
364	openclip	X

Benchmark bibtex

@article{Schiller1976,
            author = {Schiller, P. H. and Finlay, B. L. and Volman, S. F.},
            doi = {10.1152/jn.1976.39.6.1352},
            issn = {0022-3077},
            journal = {Journal of neurophysiology},
            number = {6},
            pages = {1334--1351},
            pmid = {825624},
            title = {{Quantitative studies of single-cell properties in monkey striate cortex. III. Spatial Frequency}},
            url = {http://www.ncbi.nlm.nih.gov/pubmed/825624},
            volume = {39},
            year = {1976}
            }

Ceiling

0.96.

Note that scores are relative to this ceiling.

Data: Marques2020_Schiller1976

Metric: sf_selective