privacy-masknig
This model is a fine-tuned version of distilbert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3686
- Overall Precision: 0.2885
- Overall Recall: 0.2143
- Overall F1: 0.2459
- Overall Accuracy: 0.8688
- Bod F1: 0.2375
- Building F1: 0.2871
- Cardissuer F1: 0.0
- City F1: 0.2540
- Country F1: 0.3055
- Date F1: 0.2341
- Driverlicense F1: 0.2233
- Email F1: 0.2654
- Geocoord F1: 0.1603
- Givenname1 F1: 0.2161
- Givenname2 F1: 0.1507
- Idcard F1: 0.2472
- Ip F1: 0.1851
- Lastname1 F1: 0.2296
- Lastname2 F1: 0.1305
- Lastname3 F1: 0.1245
- Pass F1: 0.1980
- Passport F1: 0.2792
- Postcode F1: 0.2794
- Secaddress F1: 0.2486
- Sex F1: 0.2933
- Socialnumber F1: 0.2258
- State F1: 0.2921
- Street F1: 0.2177
- Tel F1: 0.2409
- Time F1: 0.2893
- Title F1: 0.2814
- Username F1: 0.2368
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine_with_restarts
- lr_scheduler_warmup_ratio: 0.2
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy | Bod F1 | Building F1 | Cardissuer F1 | City F1 | Country F1 | Date F1 | Driverlicense F1 | Email F1 | Geocoord F1 | Givenname1 F1 | Givenname2 F1 | Idcard F1 | Ip F1 | Lastname1 F1 | Lastname2 F1 | Lastname3 F1 | Pass F1 | Passport F1 | Postcode F1 | Secaddress F1 | Sex F1 | Socialnumber F1 | State F1 | Street F1 | Tel F1 | Time F1 | Title F1 | Username F1 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
0.4774 | 1.0 | 62187 | 0.4611 | 0.1764 | 0.1017 | 0.1291 | 0.8380 | 0.1353 | 0.1842 | 0.0 | 0.1255 | 0.2337 | 0.1185 | 0.0936 | 0.1261 | 0.0500 | 0.0893 | 0.0506 | 0.1041 | 0.1122 | 0.1241 | 0.0463 | 0.0020 | 0.0486 | 0.1080 | 0.1726 | 0.1540 | 0.2044 | 0.0886 | 0.1588 | 0.1239 | 0.1406 | 0.1667 | 0.1583 | 0.1386 |
0.4205 | 2.0 | 124374 | 0.4272 | 0.2372 | 0.1567 | 0.1887 | 0.8542 | 0.1831 | 0.2706 | 0.0 | 0.1923 | 0.2819 | 0.1821 | 0.1521 | 0.1863 | 0.1197 | 0.0997 | 0.0662 | 0.1473 | 0.1512 | 0.1443 | 0.0955 | 0.0527 | 0.1678 | 0.1997 | 0.2469 | 0.2066 | 0.2641 | 0.1827 | 0.2266 | 0.1602 | 0.1879 | 0.2372 | 0.2202 | 0.2069 |
0.3367 | 3.0 | 186561 | 0.3686 | 0.2885 | 0.2143 | 0.2459 | 0.8688 | 0.2375 | 0.2871 | 0.0 | 0.2540 | 0.3055 | 0.2341 | 0.2233 | 0.2654 | 0.1603 | 0.2161 | 0.1507 | 0.2472 | 0.1851 | 0.2296 | 0.1305 | 0.1245 | 0.1980 | 0.2792 | 0.2794 | 0.2486 | 0.2933 | 0.2258 | 0.2921 | 0.2177 | 0.2409 | 0.2893 | 0.2814 | 0.2368 |
0.301 | 4.0 | 248748 | 0.3734 | 0.3073 | 0.2484 | 0.2747 | 0.8737 | 0.2565 | 0.3272 | 0.1429 | 0.2634 | 0.3355 | 0.2707 | 0.2591 | 0.3032 | 0.2153 | 0.2458 | 0.1847 | 0.2757 | 0.2252 | 0.2594 | 0.1680 | 0.1551 | 0.2410 | 0.3080 | 0.2945 | 0.2488 | 0.3139 | 0.2522 | 0.3007 | 0.2447 | 0.2584 | 0.3107 | 0.2933 | 0.2880 |
0.2451 | 5.0 | 310935 | 0.3895 | 0.3091 | 0.2664 | 0.2862 | 0.8744 | 0.2720 | 0.3313 | 0.0 | 0.2773 | 0.3470 | 0.2803 | 0.2732 | 0.3109 | 0.2202 | 0.2554 | 0.1945 | 0.2899 | 0.2382 | 0.2539 | 0.1800 | 0.1651 | 0.2514 | 0.3156 | 0.2982 | 0.2720 | 0.3364 | 0.2695 | 0.3196 | 0.2561 | 0.2732 | 0.3169 | 0.3054 | 0.3020 |
Framework versions
- Transformers 4.40.0
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1
- Downloads last month
- 10
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for taro-pudding/privacy-masknig
Finetuned
this model