zeroshot commited on
Commit
62617d2
1 Parent(s): 5898f13

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +394 -0
README.md CHANGED
@@ -4,6 +4,400 @@ language:
4
  - en
5
  tags:
6
  - sparse sparsity quantized onnx embeddings int8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7
  ---
8
 
9
  # bge-large-en-v1.5-quant
 
4
  - en
5
  tags:
6
  - sparse sparsity quantized onnx embeddings int8
7
+ - mteb
8
+ model-index:
9
+ - name: bge-large-en-v1.5-quant
10
+ results:
11
+ - task:
12
+ type: Classification
13
+ dataset:
14
+ type: mteb/amazon_counterfactual
15
+ name: MTEB AmazonCounterfactualClassification (en)
16
+ config: en
17
+ split: test
18
+ revision: e8379541af4e31359cca9fbcf4b00f2671dba205
19
+ metrics:
20
+ - type: accuracy
21
+ value: 75.53731343283583
22
+ - type: ap
23
+ value: 38.30609312253564
24
+ - type: f1
25
+ value: 69.42802757893695
26
+ - task:
27
+ type: STS
28
+ dataset:
29
+ type: mteb/biosses-sts
30
+ name: MTEB BIOSSES
31
+ config: default
32
+ split: test
33
+ revision: d3fb88f8f02e40887cd149695127462bbcf29b4a
34
+ metrics:
35
+ - type: cos_sim_pearson
36
+ value: 89.27346145216443
37
+ - type: cos_sim_spearman
38
+ value: 88.36526647458979
39
+ - type: euclidean_pearson
40
+ value: 86.83053354694746
41
+ - type: euclidean_spearman
42
+ value: 87.56223612880584
43
+ - type: manhattan_pearson
44
+ value: 86.59250609226758
45
+ - type: manhattan_spearman
46
+ value: 87.70681773644885
47
+ - task:
48
+ type: STS
49
+ dataset:
50
+ type: mteb/sickr-sts
51
+ name: MTEB SICK-R
52
+ config: default
53
+ split: test
54
+ revision: a6ea5a8cab320b040a23452cc28066d9beae2cee
55
+ metrics:
56
+ - type: cos_sim_pearson
57
+ value: 86.18998669716373
58
+ - type: cos_sim_spearman
59
+ value: 82.06129973984048
60
+ - type: euclidean_pearson
61
+ value: 83.65969509485801
62
+ - type: euclidean_spearman
63
+ value: 81.91666612708826
64
+ - type: manhattan_pearson
65
+ value: 83.6906794731384
66
+ - type: manhattan_spearman
67
+ value: 81.91752705367436
68
+ - task:
69
+ type: STS
70
+ dataset:
71
+ type: mteb/sts12-sts
72
+ name: MTEB STS12
73
+ config: default
74
+ split: test
75
+ revision: a0d554a64d88156834ff5ae9920b964011b16384
76
+ metrics:
77
+ - type: cos_sim_pearson
78
+ value: 86.93407086985752
79
+ - type: cos_sim_spearman
80
+ value: 78.82992283957066
81
+ - type: euclidean_pearson
82
+ value: 83.39733473832982
83
+ - type: euclidean_spearman
84
+ value: 78.86999229850214
85
+ - type: manhattan_pearson
86
+ value: 83.39397058098533
87
+ - type: manhattan_spearman
88
+ value: 78.85397971200753
89
+ - task:
90
+ type: STS
91
+ dataset:
92
+ type: mteb/sts13-sts
93
+ name: MTEB STS13
94
+ config: default
95
+ split: test
96
+ revision: 7e90230a92c190f1bf69ae9002b8cea547a64cca
97
+ metrics:
98
+ - type: cos_sim_pearson
99
+ value: 87.2586009863056
100
+ - type: cos_sim_spearman
101
+ value: 87.99415514558852
102
+ - type: euclidean_pearson
103
+ value: 86.98993652364359
104
+ - type: euclidean_spearman
105
+ value: 87.72725335668807
106
+ - type: manhattan_pearson
107
+ value: 86.897205761048
108
+ - type: manhattan_spearman
109
+ value: 87.65231103509018
110
+ - task:
111
+ type: STS
112
+ dataset:
113
+ type: mteb/sts14-sts
114
+ name: MTEB STS14
115
+ config: default
116
+ split: test
117
+ revision: 6031580fec1f6af667f0bd2da0a551cf4f0b2375
118
+ metrics:
119
+ - type: cos_sim_pearson
120
+ value: 85.41417660460755
121
+ - type: cos_sim_spearman
122
+ value: 83.50291886604928
123
+ - type: euclidean_pearson
124
+ value: 84.67758839660924
125
+ - type: euclidean_spearman
126
+ value: 83.4368059512681
127
+ - type: manhattan_pearson
128
+ value: 84.66027228213025
129
+ - type: manhattan_spearman
130
+ value: 83.43472054456252
131
+ - task:
132
+ type: STS
133
+ dataset:
134
+ type: mteb/sts15-sts
135
+ name: MTEB STS15
136
+ config: default
137
+ split: test
138
+ revision: ae752c7c21bf194d8b67fd573edf7ae58183cbe3
139
+ metrics:
140
+ - type: cos_sim_pearson
141
+ value: 88.02513262365703
142
+ - type: cos_sim_spearman
143
+ value: 89.00430907638267
144
+ - type: euclidean_pearson
145
+ value: 88.16290361497319
146
+ - type: euclidean_spearman
147
+ value: 88.6645154822661
148
+ - type: manhattan_pearson
149
+ value: 88.15337528825458
150
+ - type: manhattan_spearman
151
+ value: 88.66202950081507
152
+ - task:
153
+ type: STS
154
+ dataset:
155
+ type: mteb/sts16-sts
156
+ name: MTEB STS16
157
+ config: default
158
+ split: test
159
+ revision: 4d8694f8f0e0100860b497b999b3dbed754a0513
160
+ metrics:
161
+ - type: cos_sim_pearson
162
+ value: 85.10194022827035
163
+ - type: cos_sim_spearman
164
+ value: 86.45367112223394
165
+ - type: euclidean_pearson
166
+ value: 85.45292931769094
167
+ - type: euclidean_spearman
168
+ value: 86.06607589083283
169
+ - type: manhattan_pearson
170
+ value: 85.4111233047049
171
+ - type: manhattan_spearman
172
+ value: 86.04379654118996
173
+ - task:
174
+ type: STS
175
+ dataset:
176
+ type: mteb/sts17-crosslingual-sts
177
+ name: MTEB STS17 (en-en)
178
+ config: en-en
179
+ split: test
180
+ revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d
181
+ metrics:
182
+ - type: cos_sim_pearson
183
+ value: 89.86966589113663
184
+ - type: cos_sim_spearman
185
+ value: 89.5617056243649
186
+ - type: euclidean_pearson
187
+ value: 89.018495917952
188
+ - type: euclidean_spearman
189
+ value: 88.387335721179
190
+ - type: manhattan_pearson
191
+ value: 89.07568042943448
192
+ - type: manhattan_spearman
193
+ value: 88.51733863475219
194
+ - task:
195
+ type: STS
196
+ dataset:
197
+ type: mteb/sts22-crosslingual-sts
198
+ name: MTEB STS22 (en)
199
+ config: en
200
+ split: test
201
+ revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80
202
+ metrics:
203
+ - type: cos_sim_pearson
204
+ value: 68.38465344518238
205
+ - type: cos_sim_spearman
206
+ value: 68.15219488291783
207
+ - type: euclidean_pearson
208
+ value: 68.99169681132668
209
+ - type: euclidean_spearman
210
+ value: 68.01334641045888
211
+ - type: manhattan_pearson
212
+ value: 68.84952679202642
213
+ - type: manhattan_spearman
214
+ value: 67.85430179655137
215
+ - task:
216
+ type: STS
217
+ dataset:
218
+ type: mteb/stsbenchmark-sts
219
+ name: MTEB STSBenchmark
220
+ config: default
221
+ split: test
222
+ revision: b0fddb56ed78048fa8b90373c8a3cfc37b684831
223
+ metrics:
224
+ - type: cos_sim_pearson
225
+ value: 86.60574360222778
226
+ - type: cos_sim_spearman
227
+ value: 87.8878986593873
228
+ - type: euclidean_pearson
229
+ value: 87.11557232168404
230
+ - type: euclidean_spearman
231
+ value: 87.40944677043365
232
+ - type: manhattan_pearson
233
+ value: 87.10395398212532
234
+ - type: manhattan_spearman
235
+ value: 87.35977283466168
236
+ - task:
237
+ type: PairClassification
238
+ dataset:
239
+ type: mteb/sprintduplicatequestions-pairclassification
240
+ name: MTEB SprintDuplicateQuestions
241
+ config: default
242
+ split: test
243
+ revision: d66bd1f72af766a5cc4b0ca5e00c162f89e8cc46
244
+ metrics:
245
+ - type: cos_sim_accuracy
246
+ value: 99.84752475247525
247
+ - type: cos_sim_ap
248
+ value: 96.49316696572335
249
+ - type: cos_sim_f1
250
+ value: 92.35352532274081
251
+ - type: cos_sim_precision
252
+ value: 91.71597633136095
253
+ - type: cos_sim_recall
254
+ value: 93.0
255
+ - type: dot_accuracy
256
+ value: 99.77326732673268
257
+ - type: dot_ap
258
+ value: 93.5497681978726
259
+ - type: dot_f1
260
+ value: 88.35582208895552
261
+ - type: dot_precision
262
+ value: 88.31168831168831
263
+ - type: dot_recall
264
+ value: 88.4
265
+ - type: euclidean_accuracy
266
+ value: 99.84653465346534
267
+ - type: euclidean_ap
268
+ value: 96.36378999360083
269
+ - type: euclidean_f1
270
+ value: 92.33052944087086
271
+ - type: euclidean_precision
272
+ value: 91.38099902056807
273
+ - type: euclidean_recall
274
+ value: 93.30000000000001
275
+ - type: manhattan_accuracy
276
+ value: 99.84455445544555
277
+ - type: manhattan_ap
278
+ value: 96.36035171233175
279
+ - type: manhattan_f1
280
+ value: 92.13260761999011
281
+ - type: manhattan_precision
282
+ value: 91.1851126346719
283
+ - type: manhattan_recall
284
+ value: 93.10000000000001
285
+ - type: max_accuracy
286
+ value: 99.84752475247525
287
+ - type: max_ap
288
+ value: 96.49316696572335
289
+ - type: max_f1
290
+ value: 92.35352532274081
291
+ - task:
292
+ type: PairClassification
293
+ dataset:
294
+ type: mteb/twittersemeval2015-pairclassification
295
+ name: MTEB TwitterSemEval2015
296
+ config: default
297
+ split: test
298
+ revision: 70970daeab8776df92f5ea462b6173c0b46fd2d1
299
+ metrics:
300
+ - type: cos_sim_accuracy
301
+ value: 87.26828396018358
302
+ - type: cos_sim_ap
303
+ value: 77.79878217023162
304
+ - type: cos_sim_f1
305
+ value: 71.0425694621463
306
+ - type: cos_sim_precision
307
+ value: 68.71301775147928
308
+ - type: cos_sim_recall
309
+ value: 73.53562005277044
310
+ - type: dot_accuracy
311
+ value: 84.01978899684092
312
+ - type: dot_ap
313
+ value: 66.12134149171163
314
+ - type: dot_f1
315
+ value: 63.283507097098365
316
+ - type: dot_precision
317
+ value: 60.393191081275475
318
+ - type: dot_recall
319
+ value: 66.46437994722955
320
+ - type: euclidean_accuracy
321
+ value: 87.24444179531503
322
+ - type: euclidean_ap
323
+ value: 77.84821131946212
324
+ - type: euclidean_f1
325
+ value: 71.30456661215247
326
+ - type: euclidean_precision
327
+ value: 68.1413801394566
328
+ - type: euclidean_recall
329
+ value: 74.77572559366754
330
+ - type: manhattan_accuracy
331
+ value: 87.19079692436074
332
+ - type: manhattan_ap
333
+ value: 77.78054941055291
334
+ - type: manhattan_f1
335
+ value: 71.13002127393318
336
+ - type: manhattan_precision
337
+ value: 67.65055939062128
338
+ - type: manhattan_recall
339
+ value: 74.9868073878628
340
+ - type: max_accuracy
341
+ value: 87.26828396018358
342
+ - type: max_ap
343
+ value: 77.84821131946212
344
+ - type: max_f1
345
+ value: 71.30456661215247
346
+ - task:
347
+ type: PairClassification
348
+ dataset:
349
+ type: mteb/twitterurlcorpus-pairclassification
350
+ name: MTEB TwitterURLCorpus
351
+ config: default
352
+ split: test
353
+ revision: 8b6510b0b1fa4e4c4f879467980e9be563ec1cdf
354
+ metrics:
355
+ - type: cos_sim_accuracy
356
+ value: 88.91023402025847
357
+ - type: cos_sim_ap
358
+ value: 85.94088151184411
359
+ - type: cos_sim_f1
360
+ value: 78.25673997223645
361
+ - type: cos_sim_precision
362
+ value: 74.45433059919367
363
+ - type: cos_sim_recall
364
+ value: 82.46843239913767
365
+ - type: dot_accuracy
366
+ value: 87.91865564481701
367
+ - type: dot_ap
368
+ value: 82.75373957440969
369
+ - type: dot_f1
370
+ value: 75.97383507276201
371
+ - type: dot_precision
372
+ value: 72.67294713160854
373
+ - type: dot_recall
374
+ value: 79.5888512473052
375
+ - type: euclidean_accuracy
376
+ value: 88.8539604921023
377
+ - type: euclidean_ap
378
+ value: 85.71590936389937
379
+ - type: euclidean_f1
380
+ value: 77.82902261742242
381
+ - type: euclidean_precision
382
+ value: 74.7219270279844
383
+ - type: euclidean_recall
384
+ value: 81.20572836464429
385
+ - type: manhattan_accuracy
386
+ value: 88.78992509799356
387
+ - type: manhattan_ap
388
+ value: 85.70200619366904
389
+ - type: manhattan_f1
390
+ value: 77.85875848203065
391
+ - type: manhattan_precision
392
+ value: 72.94315506222671
393
+ - type: manhattan_recall
394
+ value: 83.48475515860795
395
+ - type: max_accuracy
396
+ value: 88.91023402025847
397
+ - type: max_ap
398
+ value: 85.94088151184411
399
+ - type: max_f1
400
+ value: 78.25673997223645
401
  ---
402
 
403
  # bge-large-en-v1.5-quant