Sample stimuli

sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9

How to use

from brainscore_vision import load_benchmark
benchmark = load_benchmark("Geirhos2021contrast-error_consistency")
score = benchmark(my_model)

Model scores

Min Alignment Max Alignment

Rank

Model

Score

1
.771
2
.731
3
.718
4
.718
5
.709
6
.708
7
.696
8
.688
9
.658
10
.647
11
.635
12
.632
13
.631
14
.623
15
.621
16
.617
17
.617
18
.606
19
.598
20
.590
21
.581
22
.577
23
.574
24
.568
25
.565
26
.563
27
.562
28
.555
29
.554
30
.547
31
.544
32
.544
33
.540
34
.533
35
.511
36
.507
37
.506
38
.503
39
.503
40
.496
41
.492
42
.490
43
.481
44
.478
45
.476
46
.460
47
.456
48
.452
49
.445
50
.433
51
.431
52
.430
53
.429
54
.418
55
.416
56
.413
57
.403
58
.400
59
.399
60
.397
61
.376
62
.370
63
.367
64
.351
65
.349
66
.347
67
.347
68
.345
69
.344
70
.340
71
.335
72
.332
73
.329
74
.328
75
.324
76
.322
77
.317
78
.309
79
.308
80
.308
81
.302
82
.301
83
.294
84
.283
85
.276
86
.274
87
.271
88
.270
89
.269
90
.265
91
.261
92
.260
93
.254
94
.253
95
.253
96
.250
97
.246
98
.242
99
.241
100
.240
101
.237
102
.229
103
.229
104
.228
105
.221
106
.221
107
.221
108
.221
109
.220
110
.219
111
.216
112
.215
113
.206
114
.204
115
.204
116
.200
117
.200
118
.193
119
.191
120
.190
121
.186
122
.184
123
.182
124
.182
125
.182
126
.181
127
.179
128
.178
129
.176
130
.176
131
.171
132
.171
133
.166
134
.166
135
.166
136
.166
137
.163
138
.158
139
.158
140
.156
141
.156
142
.156
143
.155
144
.155
145
.155
146
.155
147
.154
148
.152
149
.152
150
.151
151
.148
152
.147
153
.147
154
.147
155
.145
156
.144
157
.143
158
.140
159
.140
160
.139
161
.139
162
.135
163
.133
164
.131
165
.131
166
.125
167
.123
168
.123
169
.123
170
.123
171
.123
172
.123
173
.123
174
.123
175
.123
176
.123
177
.123
178
.123
179
.122
180
.121
181
.121
182
.120
183
.120
184
.120
185
.119
186
.119
187
.119
188
.118
189
.118
190
.117
191
.116
192
.115
193
.114
194
.113
195
.113
196
.113
197
.113
198
.113
199
.110
200
.109
201
.108
202
.108
203
.107
204
.107
205
.107
206
.106
207
.106
208
.106
209
.106
210
.105
211
.103
212
.102
213
.102
214
.101
215
.101
216
.101
217
.101
218
.101
219
.101
220
.099
221
.097
222
.097
223
.096
224
.094
225
.094
226
.094
227
.093
228
.091
229
.088
230
.088
231
.087
232
.085
233
.083
234
.081
235
.081
236
.081
237
.080
238
.079
239
.079
240
.075
241
.074
242
.072
243
.072
244
.071
245
.070
246
.070
247
.070
248
.070
249
.069
250
.069
251
.064
252
.058
253
.056
254
.054
255
.052
256
.052
257
.050
258
.050
259
.049
260
.049
261
.048
262
.047
263
.046
264
.045
265
.044
266
.044
267
.044
268
.044
269
.044
270
.043
271
.043
272
.038
273
.037
274
.035
275
.033
276
.033
277
.033
278
.032
279
.030
280
.027
281
.025
282
.025
283
.024
284
.024
285
.023
286
.022
287
.022
288
.022
289
.021
290
.020
291
.020
292
.020
293
.020
294
.019
295
.018
296
.018
297
.017
298
.017
299
.016
300
.016
301
.016
302
.015
303
.015
304
.014
305
.013
306
.013
307
.012
308
.012
309
.011
310
.008
311
.006
312
.006
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457

Benchmark bibtex

@article{geirhos2021partial,
              title={Partial success in closing the gap between human and machine vision},
              author={Geirhos, Robert and Narayanappa, Kantharaju and Mitzkus, Benjamin and Thieringer, Tizian and Bethge, Matthias and Wichmann, Felix A and Brendel, Wieland},
              journal={Advances in Neural Information Processing Systems},
              volume={34},
              year={2021},
              url={https://openreview.net/forum?id=QkljT4mrfs}
        }

Ceiling

0.44.

Note that scores are relative to this ceiling.

Data: Geirhos2021contrast

Metric: error_consistency