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("Geirhos2021sketch-error_consistency")
score = benchmark(my_model)

Model scores

Min Alignment Max Alignment

Rank

Model

Score

1
.776
2
.776
3
.746
4
.742
5
.734
6
.700
7
.699
8
.678
9
.669
10
.666
11
.666
12
.663
13
.658
14
.658
15
.630
16
.594
17
.585
18
.578
19
.576
20
.559
21
.556
22
.539
23
.512
24
.481
25
.461
26
.460
27
.452
28
.425
29
.421
30
.382
31
.336
32
.331
33
.314
34
.293
35
.287
36
.283
37
.278
38
.276
39
.274
40
.274
41
.271
42
.267
43
.253
44
.247
45
.245
46
.242
47
.242
48
.234
49
.232
50
.221
51
.220
52
.220
53
.213
54
.212
55
.207
56
.205
57
.204
58
.202
59
.198
60
.198
61
.191
62
.177
63
.172
64
.169
65
.163
66
.163
67
.161
68
.161
69
.160
70
.159
71
.157
72
.155
73
.155
74
.153
75
.153
76
.149
77
.147
78
.145
79
.143
80
.143
81
.142
82
.141
83
.139
84
.138
85
.138
86
.136
87
.135
88
.132
89
.131
90
.127
91
.127
92
.125
93
.125
94
.125
95
.124
96
.124
97
.124
98
.121
99
.120
100
.119
101
.119
102
.119
103
.118
104
.118
105
.117
106
.117
107
.116
108
.115
109
.115
110
.115
111
.114
112
.114
113
.114
114
.113
115
.112
116
.112
117
.112
118
.111
119
.111
120
.111
121
.110
122
.109
123
.107
124
.105
125
.104
126
.104
127
.103
128
.103
129
.103
130
.103
131
.102
132
.102
133
.101
134
.101
135
.099
136
.098
137
.097
138
.097
139
.097
140
.095
141
.095
142
.094
143
.093
144
.092
145
.091
146
.091
147
.090
148
.090
149
.088
150
.088
151
.088
152
.088
153
.086
154
.086
155
.084
156
.083
157
.082
158
.080
159
.078
160
.077
161
.076
162
.075
163
.075
164
.073
165
.073
166
.072
167
.072
168
.070
169
.070
170
.070
171
.069
172
.068
173
.068
174
.068
175
.068
176
.067
177
.067
178
.067
179
.066
180
.066
181
.064
182
.064
183
.064
184
.063
185
.063
186
.063
187
.062
188
.062
189
.061
190
.060
191
.058
192
.058
193
.057
194
.057
195
.056
196
.056
197
.055
198
.055
199
.055
200
.054
201
.053
202
.053
203
.053
204
.052
205
.052
206
.052
207
.051
208
.051
209
.049
210
.049
211
.049
212
.049
213
.048
214
.048
215
.047
216
.047
217
.047
218
.046
219
.046
220
.046
221
.046
222
.045
223
.045
224
.045
225
.044
226
.043
227
.043
228
.043
229
.042
230
.042
231
.042
232
.042
233
.041
234
.041
235
.041
236
.041
237
.041
238
.041
239
.041
240
.041
241
.041
242
.041
243
.041
244
.041
245
.041
246
.041
247
.040
248
.039
249
.039
250
.039
251
.039
252
.038
253
.038
254
.038
255
.036
256
.035
257
.034
258
.034
259
.032
260
.032
261
.032
262
.032
263
.032
264
.031
265
.031
266
.031
267
.030
268
.030
269
.030
270
.029
271
.028
272
.027
273
.027
274
.026
275
.026
276
.025
277
.025
278
.025
279
.024
280
.024
281
.024
282
.024
283
.024
284
.023
285
.023
286
.023
287
.023
288
.022
289
.022
290
.021
291
.020
292
.020
293
.018
294
.014
295
.014
296
.014
297
.011
298
.009
299
.005
300
.003
301
.003
302
.002
303
.002
304
.002
305
.002
306
.002
307
.001
308
309
310
311
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

0.37.

Note that scores are relative to this ceiling.

Data: Geirhos2021sketch

Metric: error_consistency