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("Geirhos2021powerequalisation-error_consistency")
score = benchmark(my_model)

Model scores

Min Alignment Max Alignment

Rank

Model

Score

1
.755
2
.718
3
.713
4
.709
5
.707
6
.707
7
.705
8
.680
9
.672
10
.670
11
.668
12
.651
13
.650
14
.647
15
.638
16
.633
17
.599
18
.592
19
.586
20
.579
21
.564
22
.560
23
.548
24
.540
25
.501
26
.494
27
.489
28
.486
29
.482
30
.472
31
.468
32
.467
33
.462
34
.455
35
.453
36
.451
37
.450
38
.427
39
.427
40
.427
41
.397
42
.388
43
.382
44
.380
45
.378
46
.363
47
.339
48
.335
49
.323
50
.317
51
.317
52
.315
53
.307
54
.295
55
.292
56
.289
57
.289
58
.287
59
.287
60
.280
61
.278
62
.269
63
.269
64
.260
65
.260
66
.257
67
.248
68
.246
69
.245
70
.237
71
.234
72
.228
73
.227
74
.222
75
.221
76
.217
77
.217
78
.216
79
.214
80
.209
81
.201
82
.197
83
.195
84
.193
85
.190
86
.189
87
.189
88
.187
89
.186
90
.186
91
.185
92
.184
93
.182
94
.181
95
.181
96
.174
97
.173
98
.172
99
.169
100
.167
101
.164
102
.159
103
.155
104
.149
105
.147
106
.145
107
.145
108
.140
109
.134
110
.133
111
.132
112
.132
113
.132
114
.128
115
.127
116
.127
117
.126
118
.125
119
.125
120
.125
121
.122
122
.121
123
.119
124
.119
125
.118
126
.116
127
.115
128
.114
129
.113
130
.112
131
.108
132
.107
133
.105
134
.105
135
.100
136
.098
137
.098
138
.098
139
.098
140
.098
141
.098
142
.097
143
.097
144
.097
145
.097
146
.097
147
.097
148
.097
149
.097
150
.097
151
.097
152
.097
153
.097
154
.097
155
.096
156
.096
157
.095
158
.094
159
.094
160
.094
161
.093
162
.091
163
.090
164
.089
165
.089
166
.089
167
.088
168
.088
169
.088
170
.086
171
.085
172
.084
173
.083
174
.083
175
.083
176
.083
177
.083
178
.083
179
.083
180
.083
181
.082
182
.082
183
.078
184
.078
185
.078
186
.076
187
.075
188
.075
189
.073
190
.070
191
.070
192
.069
193
.068
194
.065
195
.065
196
.065
197
.064
198
.064
199
.063
200
.062
201
.062
202
.062
203
.062
204
.060
205
.059
206
.059
207
.057
208
.053
209
.053
210
.053
211
.053
212
.053
213
.053
214
.052
215
.050
216
.048
217
.047
218
.047
219
.046
220
.045
221
.044
222
.043
223
.041
224
.041
225
.040
226
.040
227
.040
228
.040
229
.040
230
.040
231
.039
232
.039
233
.038
234
.038
235
.038
236
.038
237
.038
238
.037
239
.037
240
.037
241
.037
242
.036
243
.036
244
.036
245
.035
246
.035
247
.034
248
.034
249
.034
250
.033
251
.033
252
.033
253
.033
254
.032
255
.032
256
.031
257
.031
258
.031
259
.031
260
.031
261
.031
262
.030
263
.029
264
.028
265
.027
266
.027
267
.027
268
.026
269
.024
270
.024
271
.023
272
.023
273
.023
274
.021
275
.021
276
.020
277
.020
278
.019
279
.019
280
.019
281
.018
282
.018
283
.017
284
.016
285
.015
286
.015
287
.014
288
.014
289
.014
290
.014
291
.013
292
.012
293
.012
294
.011
295
.011
296
.011
297
.010
298
.010
299
.010
300
.009
301
.008
302
.007
303
.006
304
.006
305
.005
306
.004
307
.003
308
.003
309
.003
310
.002
311
.002
312
.001
313
.000
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
458
459
460

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

0.51.

Note that scores are relative to this ceiling.

Data: Geirhos2021powerequalisation

Metric: error_consistency