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("Geirhos2021colour-error_consistency")
score = benchmark(my_model)

Model scores

Min Alignment Max Alignment

Rank

Model

Score

1
.891
2
.885
3
.866
4
.815
5
.812
6
.811
7
.804
8
.800
9
.799
10
.794
11
.792
12
.788
13
.788
14
.787
15
.769
16
.769
17
.766
18
.762
19
.756
20
.755
21
.748
22
.746
23
.739
24
.739
25
.738
26
.731
27
.727
28
.721
29
.716
30
.715
31
.703
32
.696
33
.692
34
.691
35
.689
36
.685
37
.684
38
.675
39
.675
40
.670
41
.657
42
.653
43
.647
44
.640
45
.637
46
.631
47
.624
48
.619
49
.617
50
.591
51
.587
52
.587
53
.585
54
.574
55
.561
56
.560
57
.551
58
.546
59
.543
60
.541
61
.538
62
.534
63
.528
64
.521
65
.521
66
.519
67
.512
68
.510
69
.505
70
.505
71
.504
72
.499
73
.491
74
.491
75
.489
76
.488
77
.479
78
.478
79
.474
80
.474
81
.474
82
.474
83
.474
84
.472
85
.470
86
.468
87
.468
88
.466
89
.464
90
.464
91
.463
92
.462
93
.461
94
.456
95
.454
96
.450
97
.450
98
.448
99
.448
100
.448
101
.448
102
.445
103
.443
104
.442
105
.441
106
.439
107
.438
108
.432
109
.430
110
.429
111
.428
112
.427
113
.422
114
.419
115
.411
116
.406
117
.406
118
.404
119
.403
120
.403
121
.400
122
.395
123
.395
124
.391
125
.390
126
.387
127
.377
128
.373
129
.370
130
.370
131
.365
132
.363
133
.361
134
.356
135
.346
136
.344
137
.344
138
.343
139
.342
140
.341
141
.328
142
.325
143
.324
144
.322
145
.320
146
.320
147
.316
148
.314
149
.314
150
.311
151
.309
152
.300
153
.299
154
.298
155
.293
156
.290
157
.290
158
.288
159
.288
160
.288
161
.286
162
.284
163
.269
164
.268
165
.263
166
.263
167
.261
168
.260
169
.260
170
.260
171
.254
172
.253
173
.252
174
.248
175
.248
176
.246
177
.239
178
.236
179
.231
180
.228
181
.218
182
.216
183
.215
184
.214
185
.214
186
.211
187
.211
188
.188
189
.182
190
.182
191
.180
192
.179
193
.177
194
.173
195
.170
196
.168
197
.165
198
.163
199
.162
200
.161
201
.159
202
.152
203
.152
204
.150
205
.150
206
.146
207
.143
208
.143
209
.137
210
.134
211
.134
212
.134
213
.134
214
.134
215
.134
216
.134
217
.134
218
.134
219
.134
220
.134
221
.134
222
.134
223
.134
224
.131
225
.130
226
.123
227
.122
228
.120
229
.119
230
.115
231
.113
232
.111
233
.108
234
.104
235
.104
236
.104
237
.104
238
.104
239
.096
240
.090
241
.084
242
.072
243
.072
244
.071
245
.068
246
.068
247
.067
248
.065
249
.062
250
.061
251
.060
252
.060
253
.058
254
.050
255
.049
256
.049
257
.049
258
.049
259
.048
260
.047
261
.045
262
.044
263
.043
264
.043
265
.041
266
.039
267
.038
268
.038
269
.037
270
.037
271
.036
272
.036
273
.035
274
.035
275
.031
276
.030
277
.030
278
.030
279
.027
280
.027
281
.027
282
.026
283
.025
284
.025
285
.025
286
.022
287
.022
288
.020
289
.020
290
.020
291
.020
292
.020
293
.020
294
.020
295
.020
296
.019
297
.015
298
.014
299
.014
300
.012
301
.011
302
.009
303
.009
304
.009
305
.009
306
.007
307
.006
308
.004
309
.004
310
.003
311
.002
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
456
457
458

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

0.42.

Note that scores are relative to this ceiling.

Data: Geirhos2021colour

Metric: error_consistency