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("Geirhos2021eidolonIII-error_consistency")
score = benchmark(my_model)

Model scores

Min Alignment Max Alignment

Rank

Model

Score

1
.726
2
.644
3
.641
4
.638
5
.629
6
.610
7
.597
8
.597
9
.582
10
.579
11
.578
12
.574
13
.569
14
.567
15
.564
16
.561
17
.558
18
.556
19
.556
20
.554
21
.551
22
.550
23
.548
24
.546
25
.541
26
.535
27
.535
28
.535
29
.521
30
.518
31
.517
32
.511
33
.505
34
.502
35
.502
36
.501
37
.500
38
.495
39
.494
40
.490
41
.489
42
.489
43
.484
44
.480
45
.477
46
.476
47
.474
48
.473
49
.473
50
.470
51
.469
52
.468
53
.468
54
.468
55
.465
56
.464
57
.464
58
.459
59
.453
60
.453
61
.451
62
.445
63
.444
64
.443
65
.442
66
.441
67
.439
68
.438
69
.437
70
.433
71
.433
72
.432
73
.432
74
.429
75
.427
76
.424
77
.423
78
.418
79
.416
80
.415
81
.415
82
.415
83
.412
84
.412
85
.400
86
.397
87
.395
88
.393
89
.391
90
.388
91
.383
92
.380
93
.379
94
.377
95
.373
96
.369
97
.368
98
.366
99
.364
100
.363
101
.360
102
.360
103
.357
104
.356
105
.353
106
.353
107
.352
108
.350
109
.346
110
.343
111
.340
112
.334
113
.333
114
.327
115
.324
116
.320
117
.320
118
.319
119
.319
120
.319
121
.318
122
.317
123
.316
124
.315
125
.314
126
.313
127
.313
128
.312
129
.304
130
.302
131
.302
132
.299
133
.299
134
.297
135
.297
136
.296
137
.295
138
.295
139
.294
140
.294
141
.292
142
.292
143
.291
144
.291
145
.291
146
.291
147
.290
148
.290
149
.287
150
.286
151
.286
152
.285
153
.284
154
.284
155
.283
156
.282
157
.281
158
.278
159
.276
160
.274
161
.273
162
.273
163
.272
164
.271
165
.267
166
.263
167
.262
168
.258
169
.257
170
.256
171
.255
172
.252
173
.252
174
.251
175
.250
176
.247
177
.245
178
.245
179
.244
180
.242
181
.241
182
.240
183
.238
184
.233
185
.231
186
.228
187
.226
188
.221
189
.220
190
.219
191
.219
192
.219
193
.215
194
.214
195
.213
196
.212
197
.204
198
.203
199
.200
200
.196
201
.190
202
.189
203
.174
204
.173
205
.171
206
.168
207
.168
208
.167
209
.167
210
.167
211
.167
212
.167
213
.167
214
.167
215
.167
216
.167
217
.167
218
.167
219
.167
220
.162
221
.161
222
.157
223
.154
224
.154
225
.151
226
.151
227
.146
228
.145
229
.142
230
.138
231
.137
232
.136
233
.135
234
.132
235
.130
236
.129
237
.129
238
.129
239
.129
240
.129
241
.129
242
.127
243
.126
244
.125
245
.124
246
.124
247
.124
248
.123
249
.123
250
.121
251
.121
252
.117
253
.113
254
.112
255
.112
256
.112
257
.112
258
.112
259
.108
260
.105
261
.105
262
.104
263
.103
264
.101
265
.099
266
.097
267
.096
268
.095
269
.094
270
.093
271
.093
272
.093
273
.091
274
.091
275
.090
276
.090
277
.089
278
.089
279
.085
280
.081
281
.080
282
.079
283
.078
284
.076
285
.071
286
.069
287
.063
288
.062
289
.057
290
.056
291
.042
292
.040
293
.039
294
.037
295
.037
296
.037
297
.037
298
.037
299
.036
300
.034
301
.033
302
.033
303
.026
304
.023
305
.018
306
.010
307
.010
308
.007
309
.003
310
.001
311
.000
312
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

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

0.46.

Note that scores are relative to this ceiling.

Data: Geirhos2021eidolonIII

Metric: error_consistency