Hugging Face's logo

clincolnoz
/
bert-base-uncased-edos

Text Classification
Transformers
PyTorch
English
bert
Generated from Trainer
Inference Endpoints
Model card Files Community
1
clincolnoz commited on
Commit
4fb0ffe
•
1 Parent(s): 71f9e69

full precision weights

Browse files
Files changed (8) hide show
  1. README.md +41 -42
  2. all_results.json +15 -15
  3. eval_results.json +11 -11
  4. predict_results_None.txt +99 -99
  5. pytorch_model.bin +1 -1
  6. train_results.json +4 -4
  7. trainer_state.json +54 -45
  8. training_args.bin +2 -2
README.md CHANGED
@@ -6,32 +6,32 @@ metrics:
6
  - f1
7
  - accuracy
8
  model-index:
9
- - name: final-lr2e-5-bs16
10
  results: []
11
  ---
12
 
13
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
  should probably proofread and complete it, then remove this comment. -->
15
 
16
- # final-lr2e-5-bs16
17
 
18
  This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 0.4823
21
- - F1 Macro: 0.8301
22
- - F1 Weighted: 0.8772
23
- - F1: 0.7388
24
- - Accuracy: 0.8792
25
- - Confusion Matrix: [[2834 196]
26
- [ 287 683]]
27
- - Confusion Matrix Norm: [[0.93531353 0.06468647]
28
- [0.29587629 0.70412371]]
29
- - Classification Report: precision recall f1-score support
30
- 0 0.908042 0.935314 0.921476 3030.00000
31
- 1 0.777019 0.704124 0.738778 970.00000
32
- accuracy 0.879250 0.879250 0.879250 0.87925
33
- macro avg 0.842531 0.819719 0.830127 4000.00000
34
- weighted avg 0.876269 0.879250 0.877172 4000.00000
35
 
36
  ## Model description
37
 
@@ -57,36 +57,35 @@ The following hyperparameters were used during training:
57
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
58
  - lr_scheduler_type: linear
59
  - num_epochs: 3.0
60
- - mixed_precision_training: Native AMP
61
 
62
  ### Training results
63
 
64
  | Training Loss | Epoch | Step | Validation Loss | F1 Macro | F1 Weighted | F1 | Accuracy | Confusion Matrix | Confusion Matrix Norm | Classification Report |
65
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:-----------:|:------:|:--------:|:--------------------------:|:--------------------------------------------------:|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|
66
- | 0.3333 | 1.0 | 1000 | 0.3064 | 0.8165 | 0.8672 | 0.7181 | 0.8692 | [[2811 219]
67
- [ 304 666]] | [[0.92772277 0.07227723]
68
- [0.31340206 0.68659794]] | precision recall f1-score support
69
- 0 0.902408 0.927723 0.914890 3030.00000
70
- 1 0.752542 0.686598 0.718059 970.00000
71
- accuracy 0.869250 0.869250 0.869250 0.86925
72
- macro avg 0.827475 0.807160 0.816475 4000.00000
73
- weighted avg 0.866065 0.869250 0.867159 4000.00000 |
74
- | 0.2271 | 2.0 | 2000 | 0.3905 | 0.8238 | 0.8708 | 0.7326 | 0.871 | [[2777 253]
75
- [ 263 707]] | [[0.91650165 0.08349835]
76
- [0.27113402 0.72886598]] | precision recall f1-score support
77
- 0 0.913487 0.916502 0.914992 3030.000
78
- 1 0.736458 0.728866 0.732642 970.000
79
- accuracy 0.871000 0.871000 0.871000 0.871
80
- macro avg 0.824973 0.822684 0.823817 4000.000
81
- weighted avg 0.870557 0.871000 0.870772 4000.000 |
82
- | 0.1435 | 3.0 | 3000 | 0.4823 | 0.8301 | 0.8772 | 0.7388 | 0.8792 | [[2834 196]
83
- [ 287 683]] | [[0.93531353 0.06468647]
84
- [0.29587629 0.70412371]] | precision recall f1-score support
85
- 0 0.908042 0.935314 0.921476 3030.00000
86
- 1 0.777019 0.704124 0.738778 970.00000
87
- accuracy 0.879250 0.879250 0.879250 0.87925
88
- macro avg 0.842531 0.819719 0.830127 4000.00000
89
- weighted avg 0.876269 0.879250 0.877172 4000.00000 |
90
 
91
 
92
  ### Framework versions
 
6
  - f1
7
  - accuracy
8
  model-index:
9
+ - name: final-lr2e-5-bs16-fullprecision
10
  results: []
11
  ---
12
 
13
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
  should probably proofread and complete it, then remove this comment. -->
15
 
16
+ # final-lr2e-5-bs16-fullprecision
17
 
18
  This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 0.4633
21
+ - F1 Macro: 0.8276
22
+ - F1 Weighted: 0.8754
23
+ - F1: 0.7348
24
+ - Accuracy: 0.8775
25
+ - Confusion Matrix: [[2831 199]
26
+ [ 291 679]]
27
+ - Confusion Matrix Norm: [[0.93432343 0.06567657]
28
+ [0.3 0.7 ]]
29
+ - Classification Report: precision recall f1-score support
30
+ 0 0.906791 0.934323 0.920351 3030.0000
31
+ 1 0.773349 0.700000 0.734848 970.0000
32
+ accuracy 0.877500 0.877500 0.877500 0.8775
33
+ macro avg 0.840070 0.817162 0.827600 4000.0000
34
+ weighted avg 0.874431 0.877500 0.875367 4000.0000
35
 
36
  ## Model description
37
 
 
57
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
58
  - lr_scheduler_type: linear
59
  - num_epochs: 3.0
 
60
 
61
  ### Training results
62
 
63
  | Training Loss | Epoch | Step | Validation Loss | F1 Macro | F1 Weighted | F1 | Accuracy | Confusion Matrix | Confusion Matrix Norm | Classification Report |
64
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:-----------:|:------:|:--------:|:--------------------------:|:--------------------------------------------------:|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|
65
+ | 0.3362 | 1.0 | 1000 | 0.3034 | 0.8182 | 0.8693 | 0.7191 | 0.8722 | [[2835 195]
66
+ [ 316 654]] | [[0.93564356 0.06435644]
67
+ [0.3257732 0.6742268 ]] | precision recall f1-score support
68
+ 0 0.899714 0.935644 0.917327 3030.00000
69
+ 1 0.770318 0.674227 0.719076 970.00000
70
+ accuracy 0.872250 0.872250 0.872250 0.87225
71
+ macro avg 0.835016 0.804935 0.818202 4000.00000
72
+ weighted avg 0.868336 0.872250 0.869251 4000.00000 |
73
+ | 0.2352 | 2.0 | 2000 | 0.3730 | 0.8270 | 0.8730 | 0.7374 | 0.8732 | [[2781 249]
74
+ [ 258 712]] | [[0.91782178 0.08217822]
75
+ [0.26597938 0.73402062]] | precision recall f1-score support
76
+ 0 0.915104 0.917822 0.916461 3030.00000
77
+ 1 0.740895 0.734021 0.737442 970.00000
78
+ accuracy 0.873250 0.873250 0.873250 0.87325
79
+ macro avg 0.827999 0.825921 0.826951 4000.00000
80
+ weighted avg 0.872858 0.873250 0.873049 4000.00000 |
81
+ | 0.1566 | 3.0 | 3000 | 0.4633 | 0.8276 | 0.8754 | 0.7348 | 0.8775 | [[2831 199]
82
+ [ 291 679]] | [[0.93432343 0.06567657]
83
+ [0.3 0.7 ]] | precision recall f1-score support
84
+ 0 0.906791 0.934323 0.920351 3030.0000
85
+ 1 0.773349 0.700000 0.734848 970.0000
86
+ accuracy 0.877500 0.877500 0.877500 0.8775
87
+ macro avg 0.840070 0.817162 0.827600 4000.0000
88
+ weighted avg 0.874431 0.877500 0.875367 4000.0000 |
89
 
90
 
91
  ### Framework versions
all_results.json CHANGED
@@ -1,20 +1,20 @@
1
  {
2
  "epoch": 3.0,
3
- "eval_accuracy": 0.87925,
4
- "eval_classification_report": " precision recall f1-score support\n0 0.908042 0.935314 0.921476 3030.00000\n1 0.777019 0.704124 0.738778 970.00000\naccuracy 0.879250 0.879250 0.879250 0.87925\nmacro avg 0.842531 0.819719 0.830127 4000.00000\nweighted avg 0.876269 0.879250 0.877172 4000.00000",
5
- "eval_confusion_matrix": "[[2834 196]\n [ 287 683]]",
6
- "eval_confusion_matrix_norm": "[[0.93531353 0.06468647]\n [0.29587629 0.70412371]]",
7
- "eval_f1": 0.7387777176852353,
8
- "eval_f1_macro": 0.8301269502098749,
9
- "eval_f1_weighted": 0.8771718049600645,
10
- "eval_loss": 0.4823172092437744,
11
- "eval_runtime": 9.641,
12
  "eval_samples": 4000,
13
- "eval_samples_per_second": 414.893,
14
- "eval_steps_per_second": 25.931,
15
- "train_loss": 0.2520793151855469,
16
- "train_runtime": 436.5353,
17
  "train_samples": 16000,
18
- "train_samples_per_second": 109.957,
19
- "train_steps_per_second": 6.872
20
  }
 
1
  {
2
  "epoch": 3.0,
3
+ "eval_accuracy": 0.8775,
4
+ "eval_classification_report": " precision recall f1-score support\n0 0.906791 0.934323 0.920351 3030.0000\n1 0.773349 0.700000 0.734848 970.0000\naccuracy 0.877500 0.877500 0.877500 0.8775\nmacro avg 0.840070 0.817162 0.827600 4000.0000\nweighted avg 0.874431 0.877500 0.875367 4000.0000",
5
+ "eval_confusion_matrix": "[[2831 199]\n [ 291 679]]",
6
+ "eval_confusion_matrix_norm": "[[0.93432343 0.06567657]\n [0.3 0.7 ]]",
7
+ "eval_f1": 0.7348484848484848,
8
+ "eval_f1_macro": 0.8275997950900422,
9
+ "eval_f1_weighted": 0.8753667198644444,
10
+ "eval_loss": 0.4632544219493866,
11
+ "eval_runtime": 16.6824,
12
  "eval_samples": 4000,
13
+ "eval_samples_per_second": 239.773,
14
+ "eval_steps_per_second": 14.986,
15
+ "train_loss": 0.2591003138224284,
16
+ "train_runtime": 651.1299,
17
  "train_samples": 16000,
18
+ "train_samples_per_second": 73.718,
19
+ "train_steps_per_second": 4.607
20
  }
eval_results.json CHANGED
@@ -1,15 +1,15 @@
1
  {
2
  "epoch": 3.0,
3
- "eval_accuracy": 0.87925,
4
- "eval_classification_report": " precision recall f1-score support\n0 0.908042 0.935314 0.921476 3030.00000\n1 0.777019 0.704124 0.738778 970.00000\naccuracy 0.879250 0.879250 0.879250 0.87925\nmacro avg 0.842531 0.819719 0.830127 4000.00000\nweighted avg 0.876269 0.879250 0.877172 4000.00000",
5
- "eval_confusion_matrix": "[[2834 196]\n [ 287 683]]",
6
- "eval_confusion_matrix_norm": "[[0.93531353 0.06468647]\n [0.29587629 0.70412371]]",
7
- "eval_f1": 0.7387777176852353,
8
- "eval_f1_macro": 0.8301269502098749,
9
- "eval_f1_weighted": 0.8771718049600645,
10
- "eval_loss": 0.4823172092437744,
11
- "eval_runtime": 9.641,
12
  "eval_samples": 4000,
13
- "eval_samples_per_second": 414.893,
14
- "eval_steps_per_second": 25.931
15
  }
 
1
  {
2
  "epoch": 3.0,
3
+ "eval_accuracy": 0.8775,
4
+ "eval_classification_report": " precision recall f1-score support\n0 0.906791 0.934323 0.920351 3030.0000\n1 0.773349 0.700000 0.734848 970.0000\naccuracy 0.877500 0.877500 0.877500 0.8775\nmacro avg 0.840070 0.817162 0.827600 4000.0000\nweighted avg 0.874431 0.877500 0.875367 4000.0000",
5
+ "eval_confusion_matrix": "[[2831 199]\n [ 291 679]]",
6
+ "eval_confusion_matrix_norm": "[[0.93432343 0.06567657]\n [0.3 0.7 ]]",
7
+ "eval_f1": 0.7348484848484848,
8
+ "eval_f1_macro": 0.8275997950900422,
9
+ "eval_f1_weighted": 0.8753667198644444,
10
+ "eval_loss": 0.4632544219493866,
11
+ "eval_runtime": 16.6824,
12
  "eval_samples": 4000,
13
+ "eval_samples_per_second": 239.773,
14
+ "eval_steps_per_second": 14.986
15
  }
predict_results_None.txt CHANGED
@@ -33,8 +33,8 @@ index prediction
33
  31 not sexist
34
  32 not sexist
35
  33 not sexist
36
- 34 not sexist
37
- 35 not sexist
38
  36 not sexist
39
  37 not sexist
40
  38 not sexist
@@ -54,7 +54,7 @@ index prediction
54
  52 not sexist
55
  53 not sexist
56
  54 not sexist
57
- 55 sexist
58
  56 not sexist
59
  57 not sexist
60
  58 not sexist
@@ -85,7 +85,7 @@ index prediction
85
  83 not sexist
86
  84 not sexist
87
  85 sexist
88
- 86 not sexist
89
  87 sexist
90
  88 not sexist
91
  89 sexist
@@ -104,7 +104,7 @@ index prediction
104
  102 not sexist
105
  103 sexist
106
  104 not sexist
107
- 105 sexist
108
  106 not sexist
109
  107 not sexist
110
  108 not sexist
@@ -142,7 +142,7 @@ index prediction
142
  140 not sexist
143
  141 not sexist
144
  142 not sexist
145
- 143 sexist
146
  144 not sexist
147
  145 sexist
148
  146 not sexist
@@ -203,7 +203,7 @@ index prediction
203
  201 sexist
204
  202 not sexist
205
  203 not sexist
206
- 204 not sexist
207
  205 sexist
208
  206 sexist
209
  207 not sexist
@@ -223,7 +223,7 @@ index prediction
223
  221 not sexist
224
  222 not sexist
225
  223 not sexist
226
- 224 not sexist
227
  225 not sexist
228
  226 not sexist
229
  227 not sexist
@@ -255,7 +255,7 @@ index prediction
255
  253 not sexist
256
  254 not sexist
257
  255 not sexist
258
- 256 not sexist
259
  257 not sexist
260
  258 not sexist
261
  259 not sexist
@@ -275,7 +275,7 @@ index prediction
275
  273 not sexist
276
  274 not sexist
277
  275 not sexist
278
- 276 not sexist
279
  277 sexist
280
  278 not sexist
281
  279 not sexist
@@ -305,7 +305,7 @@ index prediction
305
  303 sexist
306
  304 not sexist
307
  305 not sexist
308
- 306 not sexist
309
  307 not sexist
310
  308 sexist
311
  309 not sexist
@@ -341,7 +341,7 @@ index prediction
341
  339 not sexist
342
  340 not sexist
343
  341 not sexist
344
- 342 not sexist
345
  343 not sexist
346
  344 not sexist
347
  345 not sexist
@@ -380,7 +380,7 @@ index prediction
380
  378 not sexist
381
  379 not sexist
382
  380 not sexist
383
- 381 sexist
384
  382 not sexist
385
  383 not sexist
386
  384 sexist
@@ -443,7 +443,7 @@ index prediction
443
  441 not sexist
444
  442 sexist
445
  443 not sexist
446
- 444 sexist
447
  445 sexist
448
  446 sexist
449
  447 not sexist
@@ -514,7 +514,7 @@ index prediction
514
  512 not sexist
515
  513 not sexist
516
  514 not sexist
517
- 515 sexist
518
  516 not sexist
519
  517 sexist
520
  518 not sexist
@@ -541,7 +541,7 @@ index prediction
541
  539 sexist
542
  540 not sexist
543
  541 not sexist
544
- 542 not sexist
545
  543 not sexist
546
  544 not sexist
547
  545 sexist
@@ -561,7 +561,7 @@ index prediction
561
  559 not sexist
562
  560 not sexist
563
  561 sexist
564
- 562 sexist
565
  563 not sexist
566
  564 not sexist
567
  565 sexist
@@ -581,7 +581,7 @@ index prediction
581
  579 not sexist
582
  580 sexist
583
  581 sexist
584
- 582 not sexist
585
  583 not sexist
586
  584 not sexist
587
  585 not sexist
@@ -631,16 +631,16 @@ index prediction
631
  629 not sexist
632
  630 not sexist
633
  631 not sexist
634
- 632 sexist
635
  633 not sexist
636
  634 not sexist
637
  635 sexist
638
  636 not sexist
639
  637 not sexist
640
- 638 sexist
641
  639 not sexist
642
  640 not sexist
643
- 641 sexist
644
  642 not sexist
645
  643 not sexist
646
  644 not sexist
@@ -672,7 +672,7 @@ index prediction
672
  670 not sexist
673
  671 not sexist
674
  672 not sexist
675
- 673 not sexist
676
  674 sexist
677
  675 sexist
678
  676 not sexist
@@ -701,7 +701,7 @@ index prediction
701
  699 sexist
702
  700 not sexist
703
  701 sexist
704
- 702 sexist
705
  703 not sexist
706
  704 not sexist
707
  705 not sexist
@@ -720,7 +720,7 @@ index prediction
720
  718 not sexist
721
  719 not sexist
722
  720 not sexist
723
- 721 not sexist
724
  722 not sexist
725
  723 sexist
726
  724 sexist
@@ -812,7 +812,7 @@ index prediction
812
  810 not sexist
813
  811 not sexist
814
  812 not sexist
815
- 813 sexist
816
  814 not sexist
817
  815 not sexist
818
  816 sexist
@@ -846,7 +846,7 @@ index prediction
846
  844 not sexist
847
  845 not sexist
848
  846 not sexist
849
- 847 sexist
850
  848 not sexist
851
  849 sexist
852
  850 not sexist
@@ -920,7 +920,7 @@ index prediction
920
  918 not sexist
921
  919 sexist
922
  920 sexist
923
- 921 not sexist
924
  922 sexist
925
  923 not sexist
926
  924 sexist
@@ -937,7 +937,7 @@ index prediction
937
  935 sexist
938
  936 not sexist
939
  937 not sexist
940
- 938 not sexist
941
  939 not sexist
942
  940 not sexist
943
  941 not sexist
@@ -949,7 +949,7 @@ index prediction
949
  947 not sexist
950
  948 not sexist
951
  949 sexist
952
- 950 sexist
953
  951 sexist
954
  952 not sexist
955
  953 not sexist
@@ -965,9 +965,9 @@ index prediction
965
  963 not sexist
966
  964 not sexist
967
  965 not sexist
968
- 966 sexist
969
  967 not sexist
970
- 968 sexist
971
  969 not sexist
972
  970 not sexist
973
  971 not sexist
@@ -999,7 +999,7 @@ index prediction
999
  997 sexist
1000
  998 not sexist
1001
  999 sexist
1002
- 1000 sexist
1003
  1001 sexist
1004
  1002 sexist
1005
  1003 sexist
@@ -1038,7 +1038,7 @@ index prediction
1038
  1036 sexist
1039
  1037 not sexist
1040
  1038 not sexist
1041
- 1039 not sexist
1042
  1040 not sexist
1043
  1041 sexist
1044
  1042 not sexist
@@ -1063,7 +1063,7 @@ index prediction
1063
  1061 not sexist
1064
  1062 sexist
1065
  1063 not sexist
1066
- 1064 sexist
1067
  1065 not sexist
1068
  1066 not sexist
1069
  1067 sexist
@@ -1119,7 +1119,7 @@ index prediction
1119
  1117 not sexist
1120
  1118 sexist
1121
  1119 not sexist
1122
- 1120 sexist
1123
  1121 not sexist
1124
  1122 not sexist
1125
  1123 sexist
@@ -1194,12 +1194,12 @@ index prediction
1194
  1192 not sexist
1195
  1193 not sexist
1196
  1194 not sexist
1197
- 1195 not sexist
1198
  1196 not sexist
1199
  1197 not sexist
1200
  1198 not sexist
1201
  1199 not sexist
1202
- 1200 sexist
1203
  1201 not sexist
1204
  1202 not sexist
1205
  1203 not sexist
@@ -1234,7 +1234,7 @@ index prediction
1234
  1232 not sexist
1235
  1233 not sexist
1236
  1234 not sexist
1237
- 1235 not sexist
1238
  1236 not sexist
1239
  1237 not sexist
1240
  1238 not sexist
@@ -1342,7 +1342,7 @@ index prediction
1342
  1340 not sexist
1343
  1341 not sexist
1344
  1342 sexist
1345
- 1343 not sexist
1346
  1344 not sexist
1347
  1345 not sexist
1348
  1346 not sexist
@@ -1367,12 +1367,12 @@ index prediction
1367
  1365 not sexist
1368
  1366 not sexist
1369
  1367 not sexist
1370
- 1368 not sexist
1371
  1369 sexist
1372
  1370 not sexist
1373
  1371 not sexist
1374
  1372 not sexist
1375
- 1373 not sexist
1376
  1374 not sexist
1377
  1375 not sexist
1378
  1376 not sexist
@@ -1450,7 +1450,7 @@ index prediction
1450
  1448 not sexist
1451
  1449 not sexist
1452
  1450 not sexist
1453
- 1451 not sexist
1454
  1452 sexist
1455
  1453 not sexist
1456
  1454 not sexist
@@ -1461,7 +1461,7 @@ index prediction
1461
  1459 sexist
1462
  1460 not sexist
1463
  1461 not sexist
1464
- 1462 not sexist
1465
  1463 not sexist
1466
  1464 not sexist
1467
  1465 not sexist
@@ -1487,7 +1487,7 @@ index prediction
1487
  1485 sexist
1488
  1486 not sexist
1489
  1487 sexist
1490
- 1488 not sexist
1491
  1489 not sexist
1492
  1490 sexist
1493
  1491 not sexist
@@ -1580,12 +1580,12 @@ index prediction
1580
  1578 not sexist
1581
  1579 sexist
1582
  1580 not sexist
1583
- 1581 sexist
1584
  1582 not sexist
1585
  1583 not sexist
1586
  1584 not sexist
1587
  1585 not sexist
1588
- 1586 sexist
1589
  1587 not sexist
1590
  1588 not sexist
1591
  1589 not sexist
@@ -1672,14 +1672,14 @@ index prediction
1672
  1670 sexist
1673
  1671 not sexist
1674
  1672 not sexist
1675
- 1673 not sexist
1676
  1674 not sexist
1677
  1675 not sexist
1678
  1676 sexist
1679
  1677 not sexist
1680
  1678 not sexist
1681
  1679 not sexist
1682
- 1680 sexist
1683
  1681 not sexist
1684
  1682 not sexist
1685
  1683 not sexist
@@ -1693,7 +1693,7 @@ index prediction
1693
  1691 not sexist
1694
  1692 not sexist
1695
  1693 not sexist
1696
- 1694 not sexist
1697
  1695 not sexist
1698
  1696 not sexist
1699
  1697 not sexist
@@ -1784,7 +1784,7 @@ index prediction
1784
  1782 not sexist
1785
  1783 sexist
1786
  1784 not sexist
1787
- 1785 sexist
1788
  1786 sexist
1789
  1787 sexist
1790
  1788 not sexist
@@ -1858,7 +1858,7 @@ index prediction
1858
  1856 not sexist
1859
  1857 sexist
1860
  1858 not sexist
1861
- 1859 sexist
1862
  1860 not sexist
1863
  1861 not sexist
1864
  1862 not sexist
@@ -1934,7 +1934,7 @@ index prediction
1934
  1932 not sexist
1935
  1933 sexist
1936
  1934 not sexist
1937
- 1935 not sexist
1938
  1936 not sexist
1939
  1937 not sexist
1940
  1938 not sexist
@@ -1966,7 +1966,7 @@ index prediction
1966
  1964 not sexist
1967
  1965 not sexist
1968
  1966 not sexist
1969
- 1967 not sexist
1970
  1968 not sexist
1971
  1969 sexist
1972
  1970 not sexist
@@ -1994,7 +1994,7 @@ index prediction
1994
  1992 not sexist
1995
  1993 not sexist
1996
  1994 not sexist
1997
- 1995 sexist
1998
  1996 not sexist
1999
  1997 not sexist
2000
  1998 sexist
@@ -2005,7 +2005,7 @@ index prediction
2005
  2003 not sexist
2006
  2004 not sexist
2007
  2005 sexist
2008
- 2006 sexist
2009
  2007 sexist
2010
  2008 not sexist
2011
  2009 not sexist
@@ -2014,7 +2014,7 @@ index prediction
2014
  2012 not sexist
2015
  2013 not sexist
2016
  2014 not sexist
2017
- 2015 not sexist
2018
  2016 not sexist
2019
  2017 not sexist
2020
  2018 not sexist
@@ -2034,7 +2034,7 @@ index prediction
2034
  2032 sexist
2035
  2033 not sexist
2036
  2034 not sexist
2037
- 2035 sexist
2038
  2036 not sexist
2039
  2037 sexist
2040
  2038 not sexist
@@ -2080,8 +2080,8 @@ index prediction
2080
  2078 not sexist
2081
  2079 not sexist
2082
  2080 sexist
2083
- 2081 sexist
2084
- 2082 sexist
2085
  2083 not sexist
2086
  2084 not sexist
2087
  2085 not sexist
@@ -2121,7 +2121,7 @@ index prediction
2121
  2119 sexist
2122
  2120 not sexist
2123
  2121 not sexist
2124
- 2122 not sexist
2125
  2123 sexist
2126
  2124 not sexist
2127
  2125 sexist
@@ -2266,13 +2266,13 @@ index prediction
2266
  2264 not sexist
2267
  2265 not sexist
2268
  2266 not sexist
2269
- 2267 sexist
2270
  2268 not sexist
2271
  2269 not sexist
2272
- 2270 sexist
2273
  2271 not sexist
2274
  2272 not sexist
2275
- 2273 sexist
2276
  2274 not sexist
2277
  2275 not sexist
2278
  2276 not sexist
@@ -2345,7 +2345,7 @@ index prediction
2345
  2343 not sexist
2346
  2344 sexist
2347
  2345 not sexist
2348
- 2346 not sexist
2349
  2347 not sexist
2350
  2348 not sexist
2351
  2349 not sexist
@@ -2389,7 +2389,7 @@ index prediction
2389
  2387 not sexist
2390
  2388 not sexist
2391
  2389 not sexist
2392
- 2390 not sexist
2393
  2391 not sexist
2394
  2392 not sexist
2395
  2393 sexist
@@ -2451,7 +2451,7 @@ index prediction
2451
  2449 not sexist
2452
  2450 not sexist
2453
  2451 not sexist
2454
- 2452 sexist
2455
  2453 not sexist
2456
  2454 sexist
2457
  2455 not sexist
@@ -2501,7 +2501,7 @@ index prediction
2501
  2499 sexist
2502
  2500 not sexist
2503
  2501 not sexist
2504
- 2502 not sexist
2505
  2503 not sexist
2506
  2504 not sexist
2507
  2505 sexist
@@ -2551,7 +2551,7 @@ index prediction
2551
  2549 not sexist
2552
  2550 not sexist
2553
  2551 not sexist
2554
- 2552 not sexist
2555
  2553 not sexist
2556
  2554 sexist
2557
  2555 sexist
@@ -2579,7 +2579,7 @@ index prediction
2579
  2577 not sexist
2580
  2578 not sexist
2581
  2579 sexist
2582
- 2580 not sexist
2583
  2581 sexist
2584
  2582 not sexist
2585
  2583 sexist
@@ -2623,7 +2623,7 @@ index prediction
2623
  2621 not sexist
2624
  2622 not sexist
2625
  2623 not sexist
2626
- 2624 sexist
2627
  2625 not sexist
2628
  2626 not sexist
2629
  2627 not sexist
@@ -2640,7 +2640,7 @@ index prediction
2640
  2638 not sexist
2641
  2639 sexist
2642
  2640 not sexist
2643
- 2641 not sexist
2644
  2642 not sexist
2645
  2643 not sexist
2646
  2644 sexist
@@ -2676,7 +2676,7 @@ index prediction
2676
  2674 not sexist
2677
  2675 not sexist
2678
  2676 not sexist
2679
- 2677 not sexist
2680
  2678 not sexist
2681
  2679 not sexist
2682
  2680 not sexist
@@ -2772,7 +2772,7 @@ index prediction
2772
  2770 not sexist
2773
  2771 not sexist
2774
  2772 not sexist
2775
- 2773 not sexist
2776
  2774 not sexist
2777
  2775 not sexist
2778
  2776 not sexist
@@ -2796,16 +2796,16 @@ index prediction
2796
  2794 not sexist
2797
  2795 not sexist
2798
  2796 not sexist
2799
- 2797 not sexist
2800
  2798 not sexist
2801
  2799 not sexist
2802
  2800 not sexist
2803
  2801 not sexist
2804
  2802 sexist
2805
  2803 not sexist
2806
- 2804 not sexist
2807
  2805 not sexist
2808
- 2806 not sexist
2809
  2807 not sexist
2810
  2808 not sexist
2811
  2809 not sexist
@@ -2870,7 +2870,7 @@ index prediction
2870
  2868 sexist
2871
  2869 not sexist
2872
  2870 not sexist
2873
- 2871 sexist
2874
  2872 not sexist
2875
  2873 not sexist
2876
  2874 not sexist
@@ -2969,7 +2969,7 @@ index prediction
2969
  2967 not sexist
2970
  2968 not sexist
2971
  2969 not sexist
2972
- 2970 sexist
2973
  2971 not sexist
2974
  2972 sexist
2975
  2973 sexist
@@ -2989,10 +2989,10 @@ index prediction
2989
  2987 not sexist
2990
  2988 not sexist
2991
  2989 not sexist
2992
- 2990 not sexist
2993
  2991 not sexist
2994
  2992 sexist
2995
- 2993 sexist
2996
  2994 sexist
2997
  2995 sexist
2998
  2996 not sexist
@@ -3064,7 +3064,7 @@ index prediction
3064
  3062 not sexist
3065
  3063 not sexist
3066
  3064 not sexist
3067
- 3065 sexist
3068
  3066 not sexist
3069
  3067 not sexist
3070
  3068 not sexist
@@ -3084,7 +3084,7 @@ index prediction
3084
  3082 not sexist
3085
  3083 not sexist
3086
  3084 not sexist
3087
- 3085 not sexist
3088
  3086 not sexist
3089
  3087 sexist
3090
  3088 not sexist
@@ -3189,7 +3189,7 @@ index prediction
3189
  3187 not sexist
3190
  3188 sexist
3191
  3189 sexist
3192
- 3190 sexist
3193
  3191 not sexist
3194
  3192 not sexist
3195
  3193 sexist
@@ -3259,7 +3259,7 @@ index prediction
3259
  3257 not sexist
3260
  3258 not sexist
3261
  3259 sexist
3262
- 3260 not sexist
3263
  3261 sexist
3264
  3262 sexist
3265
  3263 not sexist
@@ -3393,7 +3393,7 @@ index prediction
3393
  3391 not sexist
3394
  3392 not sexist
3395
  3393 not sexist
3396
- 3394 sexist
3397
  3395 not sexist
3398
  3396 not sexist
3399
  3397 not sexist
@@ -3411,7 +3411,7 @@ index prediction
3411
  3409 not sexist
3412
  3410 not sexist
3413
  3411 not sexist
3414
- 3412 not sexist
3415
  3413 not sexist
3416
  3414 not sexist
3417
  3415 not sexist
@@ -3423,7 +3423,7 @@ index prediction
3423
  3421 not sexist
3424
  3422 not sexist
3425
  3423 not sexist
3426
- 3424 sexist
3427
  3425 not sexist
3428
  3426 not sexist
3429
  3427 sexist
@@ -3438,7 +3438,7 @@ index prediction
3438
  3436 not sexist
3439
  3437 not sexist
3440
  3438 not sexist
3441
- 3439 sexist
3442
  3440 not sexist
3443
  3441 not sexist
3444
  3442 not sexist
@@ -3465,7 +3465,7 @@ index prediction
3465
  3463 not sexist
3466
  3464 not sexist
3467
  3465 not sexist
3468
- 3466 sexist
3469
  3467 not sexist
3470
  3468 sexist
3471
  3469 not sexist
@@ -3524,7 +3524,7 @@ index prediction
3524
  3522 not sexist
3525
  3523 not sexist
3526
  3524 not sexist
3527
- 3525 sexist
3528
  3526 not sexist
3529
  3527 not sexist
3530
  3528 not sexist
@@ -3533,7 +3533,7 @@ index prediction
3533
  3531 not sexist
3534
  3532 not sexist
3535
  3533 not sexist
3536
- 3534 sexist
3537
  3535 not sexist
3538
  3536 not sexist
3539
  3537 not sexist
@@ -3633,7 +3633,7 @@ index prediction
3633
  3631 not sexist
3634
  3632 not sexist
3635
  3633 not sexist
3636
- 3634 sexist
3637
  3635 not sexist
3638
  3636 sexist
3639
  3637 not sexist
@@ -3668,7 +3668,7 @@ index prediction
3668
  3666 not sexist
3669
  3667 not sexist
3670
  3668 not sexist
3671
- 3669 not sexist
3672
  3670 not sexist
3673
  3671 not sexist
3674
  3672 not sexist
@@ -3727,7 +3727,7 @@ index prediction
3727
  3725 not sexist
3728
  3726 not sexist
3729
  3727 not sexist
3730
- 3728 sexist
3731
  3729 not sexist
3732
  3730 not sexist
3733
  3731 sexist
@@ -3786,8 +3786,8 @@ index prediction
3786
  3784 not sexist
3787
  3785 sexist
3788
  3786 not sexist
3789
- 3787 sexist
3790
- 3788 not sexist
3791
  3789 not sexist
3792
  3790 not sexist
3793
  3791 not sexist
@@ -3903,7 +3903,7 @@ index prediction
3903
  3901 not sexist
3904
  3902 not sexist
3905
  3903 not sexist
3906
- 3904 not sexist
3907
  3905 not sexist
3908
  3906 not sexist
3909
  3907 sexist
@@ -3923,7 +3923,7 @@ index prediction
3923
  3921 not sexist
3924
  3922 not sexist
3925
  3923 not sexist
3926
- 3924 not sexist
3927
  3925 not sexist
3928
  3926 not sexist
3929
  3927 not sexist
@@ -3975,7 +3975,7 @@ index prediction
3975
  3973 not sexist
3976
  3974 not sexist
3977
  3975 not sexist
3978
- 3976 sexist
3979
  3977 not sexist
3980
  3978 not sexist
3981
  3979 not sexist
 
33
  31 not sexist
34
  32 not sexist
35
  33 not sexist
36
+ 34 sexist
37
+ 35 sexist
38
  36 not sexist
39
  37 not sexist
40
  38 not sexist
 
54
  52 not sexist
55
  53 not sexist
56
  54 not sexist
57
+ 55 not sexist
58
  56 not sexist
59
  57 not sexist
60
  58 not sexist
 
85
  83 not sexist
86
  84 not sexist
87
  85 sexist
88
+ 86 sexist
89
  87 sexist
90
  88 not sexist
91
  89 sexist
 
104
  102 not sexist
105
  103 sexist
106
  104 not sexist
107
+ 105 not sexist
108
  106 not sexist
109
  107 not sexist
110
  108 not sexist
 
142
  140 not sexist
143
  141 not sexist
144
  142 not sexist
145
+ 143 not sexist
146
  144 not sexist
147
  145 sexist
148
  146 not sexist
 
203
  201 sexist
204
  202 not sexist
205
  203 not sexist
206
+ 204 sexist
207
  205 sexist
208
  206 sexist
209
  207 not sexist
 
223
  221 not sexist
224
  222 not sexist
225
  223 not sexist
226
+ 224 sexist
227
  225 not sexist
228
  226 not sexist
229
  227 not sexist
 
255
  253 not sexist
256
  254 not sexist
257
  255 not sexist
258
+ 256 sexist
259
  257 not sexist
260
  258 not sexist
261
  259 not sexist
 
275
  273 not sexist
276
  274 not sexist
277
  275 not sexist
278
+ 276 sexist
279
  277 sexist
280
  278 not sexist
281
  279 not sexist
 
305
  303 sexist
306
  304 not sexist
307
  305 not sexist
308
+ 306 sexist
309
  307 not sexist
310
  308 sexist
311
  309 not sexist
 
341
  339 not sexist
342
  340 not sexist
343
  341 not sexist
344
+ 342 sexist
345
  343 not sexist
346
  344 not sexist
347
  345 not sexist
 
380
  378 not sexist
381
  379 not sexist
382
  380 not sexist
383
+ 381 not sexist
384
  382 not sexist
385
  383 not sexist
386
  384 sexist
 
443
  441 not sexist
444
  442 sexist
445
  443 not sexist
446
+ 444 not sexist
447
  445 sexist
448
  446 sexist
449
  447 not sexist
 
514
  512 not sexist
515
  513 not sexist
516
  514 not sexist
517
+ 515 not sexist
518
  516 not sexist
519
  517 sexist
520
  518 not sexist
 
541
  539 sexist
542
  540 not sexist
543
  541 not sexist
544
+ 542 sexist
545
  543 not sexist
546
  544 not sexist
547
  545 sexist
 
561
  559 not sexist
562
  560 not sexist
563
  561 sexist
564
+ 562 not sexist
565
  563 not sexist
566
  564 not sexist
567
  565 sexist
 
581
  579 not sexist
582
  580 sexist
583
  581 sexist
584
+ 582 sexist
585
  583 not sexist
586
  584 not sexist
587
  585 not sexist
 
631
  629 not sexist
632
  630 not sexist
633
  631 not sexist
634
+ 632 not sexist
635
  633 not sexist
636
  634 not sexist
637
  635 sexist
638
  636 not sexist
639
  637 not sexist
640
+ 638 not sexist
641
  639 not sexist
642
  640 not sexist
643
+ 641 not sexist
644
  642 not sexist
645
  643 not sexist
646
  644 not sexist
 
672
  670 not sexist
673
  671 not sexist
674
  672 not sexist
675
+ 673 sexist
676
  674 sexist
677
  675 sexist
678
  676 not sexist
 
701
  699 sexist
702
  700 not sexist
703
  701 sexist
704
+ 702 not sexist
705
  703 not sexist
706
  704 not sexist
707
  705 not sexist
 
720
  718 not sexist
721
  719 not sexist
722
  720 not sexist
723
+ 721 sexist
724
  722 not sexist
725
  723 sexist
726
  724 sexist
 
812
  810 not sexist
813
  811 not sexist
814
  812 not sexist
815
+ 813 not sexist
816
  814 not sexist
817
  815 not sexist
818
  816 sexist
 
846
  844 not sexist
847
  845 not sexist
848
  846 not sexist
849
+ 847 not sexist
850
  848 not sexist
851
  849 sexist
852
  850 not sexist
 
920
  918 not sexist
921
  919 sexist
922
  920 sexist
923
+ 921 sexist
924
  922 sexist
925
  923 not sexist
926
  924 sexist
 
937
  935 sexist
938
  936 not sexist
939
  937 not sexist
940
+ 938 sexist
941
  939 not sexist
942
  940 not sexist
943
  941 not sexist
 
949
  947 not sexist
950
  948 not sexist
951
  949 sexist
952
+ 950 not sexist
953
  951 sexist
954
  952 not sexist
955
  953 not sexist
 
965
  963 not sexist
966
  964 not sexist
967
  965 not sexist
968
+ 966 not sexist
969
  967 not sexist
970
+ 968 not sexist
971
  969 not sexist
972
  970 not sexist
973
  971 not sexist
 
999
  997 sexist
1000
  998 not sexist
1001
  999 sexist
1002
+ 1000 not sexist
1003
  1001 sexist
1004
  1002 sexist
1005
  1003 sexist
 
1038
  1036 sexist
1039
  1037 not sexist
1040
  1038 not sexist
1041
+ 1039 sexist
1042
  1040 not sexist
1043
  1041 sexist
1044
  1042 not sexist
 
1063
  1061 not sexist
1064
  1062 sexist
1065
  1063 not sexist
1066
+ 1064 not sexist
1067
  1065 not sexist
1068
  1066 not sexist
1069
  1067 sexist
 
1119
  1117 not sexist
1120
  1118 sexist
1121
  1119 not sexist
1122
+ 1120 not sexist
1123
  1121 not sexist
1124
  1122 not sexist
1125
  1123 sexist
 
1194
  1192 not sexist
1195
  1193 not sexist
1196
  1194 not sexist
1197
+ 1195 sexist
1198
  1196 not sexist
1199
  1197 not sexist
1200
  1198 not sexist
1201
  1199 not sexist
1202
+ 1200 not sexist
1203
  1201 not sexist
1204
  1202 not sexist
1205
  1203 not sexist
 
1234
  1232 not sexist
1235
  1233 not sexist
1236
  1234 not sexist
1237
+ 1235 sexist
1238
  1236 not sexist
1239
  1237 not sexist
1240
  1238 not sexist
 
1342
  1340 not sexist
1343
  1341 not sexist
1344
  1342 sexist
1345
+ 1343 sexist
1346
  1344 not sexist
1347
  1345 not sexist
1348
  1346 not sexist
 
1367
  1365 not sexist
1368
  1366 not sexist
1369
  1367 not sexist
1370
+ 1368 sexist
1371
  1369 sexist
1372
  1370 not sexist
1373
  1371 not sexist
1374
  1372 not sexist
1375
+ 1373 sexist
1376
  1374 not sexist
1377
  1375 not sexist
1378
  1376 not sexist
 
1450
  1448 not sexist
1451
  1449 not sexist
1452
  1450 not sexist
1453
+ 1451 sexist
1454
  1452 sexist
1455
  1453 not sexist
1456
  1454 not sexist
 
1461
  1459 sexist
1462
  1460 not sexist
1463
  1461 not sexist
1464
+ 1462 sexist
1465
  1463 not sexist
1466
  1464 not sexist
1467
  1465 not sexist
 
1487
  1485 sexist
1488
  1486 not sexist
1489
  1487 sexist
1490
+ 1488 sexist
1491
  1489 not sexist
1492
  1490 sexist
1493
  1491 not sexist
 
1580
  1578 not sexist
1581
  1579 sexist
1582
  1580 not sexist
1583
+ 1581 not sexist
1584
  1582 not sexist
1585
  1583 not sexist
1586
  1584 not sexist
1587
  1585 not sexist
1588
+ 1586 not sexist
1589
  1587 not sexist
1590
  1588 not sexist
1591
  1589 not sexist
 
1672
  1670 sexist
1673
  1671 not sexist
1674
  1672 not sexist
1675
+ 1673 sexist
1676
  1674 not sexist
1677
  1675 not sexist
1678
  1676 sexist
1679
  1677 not sexist
1680
  1678 not sexist
1681
  1679 not sexist
1682
+ 1680 not sexist
1683
  1681 not sexist
1684
  1682 not sexist
1685
  1683 not sexist
 
1693
  1691 not sexist
1694
  1692 not sexist
1695
  1693 not sexist
1696
+ 1694 sexist
1697
  1695 not sexist
1698
  1696 not sexist
1699
  1697 not sexist
 
1784
  1782 not sexist
1785
  1783 sexist
1786
  1784 not sexist
1787
+ 1785 not sexist
1788
  1786 sexist
1789
  1787 sexist
1790
  1788 not sexist
 
1858
  1856 not sexist
1859
  1857 sexist
1860
  1858 not sexist
1861
+ 1859 not sexist
1862
  1860 not sexist
1863
  1861 not sexist
1864
  1862 not sexist
 
1934
  1932 not sexist
1935
  1933 sexist
1936
  1934 not sexist
1937
+ 1935 sexist
1938
  1936 not sexist
1939
  1937 not sexist
1940
  1938 not sexist
 
1966
  1964 not sexist
1967
  1965 not sexist
1968
  1966 not sexist
1969
+ 1967 sexist
1970
  1968 not sexist
1971
  1969 sexist
1972
  1970 not sexist
 
1994
  1992 not sexist
1995
  1993 not sexist
1996
  1994 not sexist
1997
+ 1995 not sexist
1998
  1996 not sexist
1999
  1997 not sexist
2000
  1998 sexist
 
2005
  2003 not sexist
2006
  2004 not sexist
2007
  2005 sexist
2008
+ 2006 not sexist
2009
  2007 sexist
2010
  2008 not sexist
2011
  2009 not sexist
 
2014
  2012 not sexist
2015
  2013 not sexist
2016
  2014 not sexist
2017
+ 2015 sexist
2018
  2016 not sexist
2019
  2017 not sexist
2020
  2018 not sexist
 
2034
  2032 sexist
2035
  2033 not sexist
2036
  2034 not sexist
2037
+ 2035 not sexist
2038
  2036 not sexist
2039
  2037 sexist
2040
  2038 not sexist
 
2080
  2078 not sexist
2081
  2079 not sexist
2082
  2080 sexist
2083
+ 2081 not sexist
2084
+ 2082 not sexist
2085
  2083 not sexist
2086
  2084 not sexist
2087
  2085 not sexist
 
2121
  2119 sexist
2122
  2120 not sexist
2123
  2121 not sexist
2124
+ 2122 sexist
2125
  2123 sexist
2126
  2124 not sexist
2127
  2125 sexist
 
2266
  2264 not sexist
2267
  2265 not sexist
2268
  2266 not sexist
2269
+ 2267 not sexist
2270
  2268 not sexist
2271
  2269 not sexist
2272
+ 2270 not sexist
2273
  2271 not sexist
2274
  2272 not sexist
2275
+ 2273 not sexist
2276
  2274 not sexist
2277
  2275 not sexist
2278
  2276 not sexist
 
2345
  2343 not sexist
2346
  2344 sexist
2347
  2345 not sexist
2348
+ 2346 sexist
2349
  2347 not sexist
2350
  2348 not sexist
2351
  2349 not sexist
 
2389
  2387 not sexist
2390
  2388 not sexist
2391
  2389 not sexist
2392
+ 2390 sexist
2393
  2391 not sexist
2394
  2392 not sexist
2395
  2393 sexist
 
2451
  2449 not sexist
2452
  2450 not sexist
2453
  2451 not sexist
2454
+ 2452 not sexist
2455
  2453 not sexist
2456
  2454 sexist
2457
  2455 not sexist
 
2501
  2499 sexist
2502
  2500 not sexist
2503
  2501 not sexist
2504
+ 2502 sexist
2505
  2503 not sexist
2506
  2504 not sexist
2507
  2505 sexist
 
2551
  2549 not sexist
2552
  2550 not sexist
2553
  2551 not sexist
2554
+ 2552 sexist
2555
  2553 not sexist
2556
  2554 sexist
2557
  2555 sexist
 
2579
  2577 not sexist
2580
  2578 not sexist
2581
  2579 sexist
2582
+ 2580 sexist
2583
  2581 sexist
2584
  2582 not sexist
2585
  2583 sexist
 
2623
  2621 not sexist
2624
  2622 not sexist
2625
  2623 not sexist
2626
+ 2624 not sexist
2627
  2625 not sexist
2628
  2626 not sexist
2629
  2627 not sexist
 
2640
  2638 not sexist
2641
  2639 sexist
2642
  2640 not sexist
2643
+ 2641 sexist
2644
  2642 not sexist
2645
  2643 not sexist
2646
  2644 sexist
 
2676
  2674 not sexist
2677
  2675 not sexist
2678
  2676 not sexist
2679
+ 2677 sexist
2680
  2678 not sexist
2681
  2679 not sexist
2682
  2680 not sexist
 
2772
  2770 not sexist
2773
  2771 not sexist
2774
  2772 not sexist
2775
+ 2773 sexist
2776
  2774 not sexist
2777
  2775 not sexist
2778
  2776 not sexist
 
2796
  2794 not sexist
2797
  2795 not sexist
2798
  2796 not sexist
2799
+ 2797 sexist
2800
  2798 not sexist
2801
  2799 not sexist
2802
  2800 not sexist
2803
  2801 not sexist
2804
  2802 sexist
2805
  2803 not sexist
2806
+ 2804 sexist
2807
  2805 not sexist
2808
+ 2806 sexist
2809
  2807 not sexist
2810
  2808 not sexist
2811
  2809 not sexist
 
2870
  2868 sexist
2871
  2869 not sexist
2872
  2870 not sexist
2873
+ 2871 not sexist
2874
  2872 not sexist
2875
  2873 not sexist
2876
  2874 not sexist
 
2969
  2967 not sexist
2970
  2968 not sexist
2971
  2969 not sexist
2972
+ 2970 not sexist
2973
  2971 not sexist
2974
  2972 sexist
2975
  2973 sexist
 
2989
  2987 not sexist
2990
  2988 not sexist
2991
  2989 not sexist
2992
+ 2990 sexist
2993
  2991 not sexist
2994
  2992 sexist
2995
+ 2993 not sexist
2996
  2994 sexist
2997
  2995 sexist
2998
  2996 not sexist
 
3064
  3062 not sexist
3065
  3063 not sexist
3066
  3064 not sexist
3067
+ 3065 not sexist
3068
  3066 not sexist
3069
  3067 not sexist
3070
  3068 not sexist
 
3084
  3082 not sexist
3085
  3083 not sexist
3086
  3084 not sexist
3087
+ 3085 sexist
3088
  3086 not sexist
3089
  3087 sexist
3090
  3088 not sexist
 
3189
  3187 not sexist
3190
  3188 sexist
3191
  3189 sexist
3192
+ 3190 not sexist
3193
  3191 not sexist
3194
  3192 not sexist
3195
  3193 sexist
 
3259
  3257 not sexist
3260
  3258 not sexist
3261
  3259 sexist
3262
+ 3260 sexist
3263
  3261 sexist
3264
  3262 sexist
3265
  3263 not sexist
 
3393
  3391 not sexist
3394
  3392 not sexist
3395
  3393 not sexist
3396
+ 3394 not sexist
3397
  3395 not sexist
3398
  3396 not sexist
3399
  3397 not sexist
 
3411
  3409 not sexist
3412
  3410 not sexist
3413
  3411 not sexist
3414
+ 3412 sexist
3415
  3413 not sexist
3416
  3414 not sexist
3417
  3415 not sexist
 
3423
  3421 not sexist
3424
  3422 not sexist
3425
  3423 not sexist
3426
+ 3424 not sexist
3427
  3425 not sexist
3428
  3426 not sexist
3429
  3427 sexist
 
3438
  3436 not sexist
3439
  3437 not sexist
3440
  3438 not sexist
3441
+ 3439 not sexist
3442
  3440 not sexist
3443
  3441 not sexist
3444
  3442 not sexist
 
3465
  3463 not sexist
3466
  3464 not sexist
3467
  3465 not sexist
3468
+ 3466 not sexist
3469
  3467 not sexist
3470
  3468 sexist
3471
  3469 not sexist
 
3524
  3522 not sexist
3525
  3523 not sexist
3526
  3524 not sexist
3527
+ 3525 not sexist
3528
  3526 not sexist
3529
  3527 not sexist
3530
  3528 not sexist
 
3533
  3531 not sexist
3534
  3532 not sexist
3535
  3533 not sexist
3536
+ 3534 not sexist
3537
  3535 not sexist
3538
  3536 not sexist
3539
  3537 not sexist
 
3633
  3631 not sexist
3634
  3632 not sexist
3635
  3633 not sexist
3636
+ 3634 not sexist
3637
  3635 not sexist
3638
  3636 sexist
3639
  3637 not sexist
 
3668
  3666 not sexist
3669
  3667 not sexist
3670
  3668 not sexist
3671
+ 3669 sexist
3672
  3670 not sexist
3673
  3671 not sexist
3674
  3672 not sexist
 
3727
  3725 not sexist
3728
  3726 not sexist
3729
  3727 not sexist
3730
+ 3728 not sexist
3731
  3729 not sexist
3732
  3730 not sexist
3733
  3731 sexist
 
3786
  3784 not sexist
3787
  3785 sexist
3788
  3786 not sexist
3789
+ 3787 not sexist
3790
+ 3788 sexist
3791
  3789 not sexist
3792
  3790 not sexist
3793
  3791 not sexist
 
3903
  3901 not sexist
3904
  3902 not sexist
3905
  3903 not sexist
3906
+ 3904 sexist
3907
  3905 not sexist
3908
  3906 not sexist
3909
  3907 sexist
 
3923
  3921 not sexist
3924
  3922 not sexist
3925
  3923 not sexist
3926
+ 3924 sexist
3927
  3925 not sexist
3928
  3926 not sexist
3929
  3927 not sexist
 
3975
  3973 not sexist
3976
  3974 not sexist
3977
  3975 not sexist
3978
+ 3976 not sexist
3979
  3977 not sexist
3980
  3978 not sexist
3981
  3979 not sexist
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4e9f8d462d8d9118a72bd13f278e2a7f6f042f093ccece0b57016b19572f8c56
3
  size 438007925
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:21b0d3a5550e58b7ef30c0020bf27b3cb40aa1df1936fc53f798b5f45ef0f41b
3
  size 438007925
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "epoch": 3.0,
3
- "train_loss": 0.2520793151855469,
4
- "train_runtime": 436.5353,
5
  "train_samples": 16000,
6
- "train_samples_per_second": 109.957,
7
- "train_steps_per_second": 6.872
8
  }
 
1
  {
2
  "epoch": 3.0,
3
+ "train_loss": 0.2591003138224284,
4
+ "train_runtime": 651.1299,
5
  "train_samples": 16000,
6
+ "train_samples_per_second": 73.718,
7
+ "train_steps_per_second": 4.607
8
  }
trainer_state.json CHANGED
@@ -9,84 +9,93 @@
9
  "log_history": [
10
  {
11
  "epoch": 0.5,
12
- "learning_rate": 1.6673333333333335e-05,
13
- "loss": 0.4114,
14
  "step": 500
15
  },
16
  {
17
  "epoch": 1.0,
18
- "learning_rate": 1.3340000000000001e-05,
19
- "loss": 0.3333,
20
  "step": 1000
21
  },
22
  {
23
  "epoch": 1.0,
24
- "eval_accuracy": 0.86925,
25
- "eval_classification_report": " precision recall f1-score support\n0 0.902408 0.927723 0.914890 3030.00000\n1 0.752542 0.686598 0.718059 970.00000\naccuracy 0.869250 0.869250 0.869250 0.86925\nmacro avg 0.827475 0.807160 0.816475 4000.00000\nweighted avg 0.866065 0.869250 0.867159 4000.00000",
26
- "eval_confusion_matrix": "[[2811 219]\n [ 304 666]]",
27
- "eval_confusion_matrix_norm": "[[0.92772277 0.07227723]\n [0.31340206 0.68659794]]",
28
- "eval_f1": 0.7180592991913748,
29
- "eval_f1_macro": 0.8164747268943042,
30
- "eval_f1_weighted": 0.8671586721613128,
31
- "eval_loss": 0.3063889145851135,
32
- "eval_runtime": 9.6345,
33
- "eval_samples_per_second": 415.174,
34
- "eval_steps_per_second": 25.948,
35
  "step": 1000
36
  },
37
  {
38
  "epoch": 1.5,
39
- "learning_rate": 1.0013333333333335e-05,
40
- "loss": 0.2455,
41
  "step": 1500
42
  },
43
  {
44
  "epoch": 2.0,
45
- "learning_rate": 6.6866666666666665e-06,
46
- "loss": 0.2271,
47
  "step": 2000
48
  },
49
  {
50
  "epoch": 2.0,
51
- "eval_accuracy": 0.871,
52
- "eval_classification_report": " precision recall f1-score support\n0 0.913487 0.916502 0.914992 3030.000\n1 0.736458 0.728866 0.732642 970.000\naccuracy 0.871000 0.871000 0.871000 0.871\nmacro avg 0.824973 0.822684 0.823817 4000.000\nweighted avg 0.870557 0.871000 0.870772 4000.000",
53
- "eval_confusion_matrix": "[[2777 253]\n [ 263 707]]",
54
- "eval_confusion_matrix_norm": "[[0.91650165 0.08349835]\n [0.27113402 0.72886598]]",
55
- "eval_f1": 0.732642487046632,
56
- "eval_f1_macro": 0.8238171249071711,
57
- "eval_f1_weighted": 0.8707720634053486,
58
- "eval_loss": 0.3905148506164551,
59
- "eval_runtime": 9.6074,
60
- "eval_samples_per_second": 416.348,
61
- "eval_steps_per_second": 26.022,
62
  "step": 2000
63
  },
64
  {
65
  "epoch": 2.5,
66
- "learning_rate": 3.3533333333333336e-06,
67
- "loss": 0.1517,
68
  "step": 2500
69
  },
70
  {
71
  "epoch": 3.0,
72
- "learning_rate": 2e-08,
73
- "loss": 0.1435,
74
  "step": 3000
75
  },
76
  {
77
  "epoch": 3.0,
78
- "eval_accuracy": 0.87925,
79
- "eval_classification_report": " precision recall f1-score support\n0 0.908042 0.935314 0.921476 3030.00000\n1 0.777019 0.704124 0.738778 970.00000\naccuracy 0.879250 0.879250 0.879250 0.87925\nmacro avg 0.842531 0.819719 0.830127 4000.00000\nweighted avg 0.876269 0.879250 0.877172 4000.00000",
80
- "eval_confusion_matrix": "[[2834 196]\n [ 287 683]]",
81
- "eval_confusion_matrix_norm": "[[0.93531353 0.06468647]\n [0.29587629 0.70412371]]",
82
- "eval_f1": 0.7387777176852353,
83
- "eval_f1_macro": 0.8301269502098749,
84
- "eval_f1_weighted": 0.8771718049600645,
85
- "eval_loss": 0.4823172092437744,
86
- "eval_runtime": 9.6225,
87
- "eval_samples_per_second": 415.692,
88
- "eval_steps_per_second": 25.981,
89
  "step": 3000
 
 
 
 
 
 
 
 
 
90
  }
91
  ],
92
  "max_steps": 3000,
 
9
  "log_history": [
10
  {
11
  "epoch": 0.5,
12
+ "learning_rate": 1.6666666666666667e-05,
13
+ "loss": 0.4103,
14
  "step": 500
15
  },
16
  {
17
  "epoch": 1.0,
18
+ "learning_rate": 1.3333333333333333e-05,
19
+ "loss": 0.3362,
20
  "step": 1000
21
  },
22
  {
23
  "epoch": 1.0,
24
+ "eval_accuracy": 0.87225,
25
+ "eval_classification_report": " precision recall f1-score support\n0 0.899714 0.935644 0.917327 3030.00000\n1 0.770318 0.674227 0.719076 970.00000\naccuracy 0.872250 0.872250 0.872250 0.87225\nmacro avg 0.835016 0.804935 0.818202 4000.00000\nweighted avg 0.868336 0.872250 0.869251 4000.00000",
26
+ "eval_confusion_matrix": "[[2835 195]\n [ 316 654]]",
27
+ "eval_confusion_matrix_norm": "[[0.93564356 0.06435644]\n [0.3257732 0.6742268 ]]",
28
+ "eval_f1": 0.7190764156129743,
29
+ "eval_f1_macro": 0.8182018544656038,
30
+ "eval_f1_weighted": 0.8692514554747078,
31
+ "eval_loss": 0.3033996522426605,
32
+ "eval_runtime": 16.7014,
33
+ "eval_samples_per_second": 239.501,
34
+ "eval_steps_per_second": 14.969,
35
  "step": 1000
36
  },
37
  {
38
  "epoch": 1.5,
39
+ "learning_rate": 1e-05,
40
+ "loss": 0.2538,
41
  "step": 1500
42
  },
43
  {
44
  "epoch": 2.0,
45
+ "learning_rate": 6.666666666666667e-06,
46
+ "loss": 0.2352,
47
  "step": 2000
48
  },
49
  {
50
  "epoch": 2.0,
51
+ "eval_accuracy": 0.87325,
52
+ "eval_classification_report": " precision recall f1-score support\n0 0.915104 0.917822 0.916461 3030.00000\n1 0.740895 0.734021 0.737442 970.00000\naccuracy 0.873250 0.873250 0.873250 0.87325\nmacro avg 0.827999 0.825921 0.826951 4000.00000\nweighted avg 0.872858 0.873250 0.873049 4000.00000",
53
+ "eval_confusion_matrix": "[[2781 249]\n [ 258 712]]",
54
+ "eval_confusion_matrix_norm": "[[0.91782178 0.08217822]\n [0.26597938 0.73402062]]",
55
+ "eval_f1": 0.737441740031072,
56
+ "eval_f1_macro": 0.826951220979451,
57
+ "eval_f1_weighted": 0.8730486036678663,
58
+ "eval_loss": 0.37301740050315857,
59
+ "eval_runtime": 16.7066,
60
+ "eval_samples_per_second": 239.426,
61
+ "eval_steps_per_second": 14.964,
62
  "step": 2000
63
  },
64
  {
65
  "epoch": 2.5,
66
+ "learning_rate": 3.3333333333333333e-06,
67
+ "loss": 0.1625,
68
  "step": 2500
69
  },
70
  {
71
  "epoch": 3.0,
72
+ "learning_rate": 0.0,
73
+ "loss": 0.1566,
74
  "step": 3000
75
  },
76
  {
77
  "epoch": 3.0,
78
+ "eval_accuracy": 0.8775,
79
+ "eval_classification_report": " precision recall f1-score support\n0 0.906791 0.934323 0.920351 3030.0000\n1 0.773349 0.700000 0.734848 970.0000\naccuracy 0.877500 0.877500 0.877500 0.8775\nmacro avg 0.840070 0.817162 0.827600 4000.0000\nweighted avg 0.874431 0.877500 0.875367 4000.0000",
80
+ "eval_confusion_matrix": "[[2831 199]\n [ 291 679]]",
81
+ "eval_confusion_matrix_norm": "[[0.93432343 0.06567657]\n [0.3 0.7 ]]",
82
+ "eval_f1": 0.7348484848484848,
83
+ "eval_f1_macro": 0.8275997950900422,
84
+ "eval_f1_weighted": 0.8753667198644444,
85
+ "eval_loss": 0.4632544219493866,
86
+ "eval_runtime": 16.6967,
87
+ "eval_samples_per_second": 239.568,
88
+ "eval_steps_per_second": 14.973,
89
  "step": 3000
90
+ },
91
+ {
92
+ "epoch": 3.0,
93
+ "step": 3000,
94
+ "total_flos": 1.262933065728e+16,
95
+ "train_loss": 0.2591003138224284,
96
+ "train_runtime": 651.1299,
97
+ "train_samples_per_second": 73.718,
98
+ "train_steps_per_second": 4.607
99
  }
100
  ],
101
  "max_steps": 3000,
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3339ea6cd4c40a8cccf32dfd823bcd135cf0d42f2a8369a143f9bc5aab5ebb78
3
- size 3579
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23bdb167d6ed64d8a4fa58d65ab66a6eb55ed2e97a0cb401fb31212f13e2e697
3
+ size 3643