Hugging Face's logo

haryoaw
/
scenario-kd-scr-ner-full_data-univner_full44

Transformers
Safetensors
xlm-roberta
Generated from Trainer
Inference Endpoints
Model card Files Community
scenario-kd-scr-ner-full_data-univner_full44 / README.md
haryoaw's picture
haryoaw
Upload tokenizer
e0f27ed verified
preview code
|
raw
history blame
6.43 kB
metadata
base_model: haryoaw/scenario-TCR-NER_data-univner_half
library_name: transformers
license: mit
metrics:
  - precision
  - recall
  - f1
  - accuracy
tags:
  - generated_from_trainer
model-index:
  - name: scenario-kd-scr-ner-full_data-univner_full44
    results: []

scenario-kd-scr-ner-full_data-univner_full44

This model is a fine-tuned version of haryoaw/scenario-TCR-NER_data-univner_half on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 1.6199
  • Precision: 0.4352
  • Recall: 0.3701
  • F1: 0.4000
  • Accuracy: 0.9387

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: 3e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 44
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 30

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
2.9268 0.5828 500 2.5728 0.4130 0.0082 0.0161 0.9245
2.2111 1.1655 1000 2.6038 0.2535 0.0757 0.1166 0.9230
1.9667 1.7483 1500 2.4899 0.2074 0.1753 0.1900 0.9168
1.7467 2.3310 2000 2.1434 0.3184 0.1717 0.2231 0.9291
1.6429 2.9138 2500 2.1913 0.2798 0.1997 0.2330 0.9274
1.4992 3.4965 3000 2.0144 0.2988 0.2192 0.2529 0.9291
1.3977 4.0793 3500 2.0470 0.3052 0.2575 0.2793 0.9284
1.2778 4.6620 4000 2.1220 0.3168 0.2727 0.2931 0.9248
1.2224 5.2448 4500 1.9196 0.3273 0.2679 0.2946 0.9303
1.1262 5.8275 5000 1.9602 0.3120 0.3111 0.3115 0.9280
1.0371 6.4103 5500 2.0035 0.3100 0.3189 0.3144 0.9238
1.0053 6.9930 6000 1.8674 0.3395 0.2821 0.3081 0.9303
0.9232 7.5758 6500 1.8771 0.3290 0.3262 0.3276 0.9303
0.872 8.1585 7000 1.8623 0.3228 0.3269 0.3249 0.9286
0.8387 8.7413 7500 1.8017 0.3676 0.3295 0.3475 0.9341
0.7799 9.3240 8000 1.9406 0.2966 0.3473 0.3200 0.9220
0.7627 9.9068 8500 1.8042 0.3618 0.3474 0.3545 0.9336
0.7129 10.4895 9000 1.7746 0.3609 0.3440 0.3522 0.9355
0.6964 11.0723 9500 1.7343 0.4023 0.3438 0.3708 0.9379
0.6547 11.6550 10000 1.7256 0.3996 0.3616 0.3796 0.9366
0.6363 12.2378 10500 1.7899 0.3701 0.3735 0.3718 0.9319
0.6183 12.8205 11000 1.8503 0.3564 0.3575 0.3570 0.9280
0.5881 13.4033 11500 1.7546 0.3679 0.3708 0.3693 0.9325
0.5822 13.9860 12000 1.6888 0.4090 0.3331 0.3672 0.9371
0.5475 14.5688 12500 1.6986 0.4197 0.3507 0.3821 0.9371
0.5396 15.1515 13000 1.7398 0.3979 0.3698 0.3833 0.9344
0.5248 15.7343 13500 1.7333 0.3914 0.3620 0.3761 0.9340
0.5096 16.3170 14000 1.6605 0.4354 0.3561 0.3918 0.9388
0.5037 16.8998 14500 1.7022 0.3882 0.3771 0.3826 0.9355
0.4839 17.4825 15000 1.6857 0.4071 0.3587 0.3814 0.9365
0.4769 18.0653 15500 1.6599 0.4432 0.3516 0.3921 0.9389
0.4607 18.6480 16000 1.6403 0.4445 0.3650 0.4009 0.9396
0.4567 19.2308 16500 1.6463 0.4321 0.3546 0.3895 0.9388
0.4449 19.8135 17000 1.6771 0.4148 0.3836 0.3986 0.9366
0.4363 20.3963 17500 1.7157 0.3993 0.3735 0.3860 0.9341
0.437 20.9790 18000 1.6571 0.4148 0.3738 0.3932 0.9372
0.4221 21.5618 18500 1.6544 0.4196 0.3582 0.3865 0.9372
0.4168 22.1445 19000 1.6168 0.4472 0.3428 0.3881 0.9399
0.409 22.7273 19500 1.6285 0.4335 0.3572 0.3917 0.9388
0.4081 23.3100 20000 1.6653 0.4058 0.3758 0.3903 0.9353
0.4009 23.8928 20500 1.6389 0.4263 0.3662 0.3940 0.9380
0.3886 24.4755 21000 1.6027 0.4632 0.3657 0.4087 0.9407
0.3947 25.0583 21500 1.6297 0.4377 0.3632 0.3969 0.9387
0.3839 25.6410 22000 1.6285 0.4317 0.3670 0.3968 0.9384
0.382 26.2238 22500 1.6517 0.4226 0.3740 0.3968 0.9372
0.3797 26.8065 23000 1.6248 0.4441 0.3711 0.4043 0.9389
0.3748 27.3893 23500 1.6254 0.4379 0.3754 0.4043 0.9388
0.3736 27.9720 24000 1.6162 0.4515 0.3659 0.4042 0.9392
0.3714 28.5548 24500 1.6312 0.4289 0.3748 0.4001 0.9380
0.3719 29.1375 25000 1.6199 0.4390 0.3720 0.4027 0.9384
0.3684 29.7203 25500 1.6199 0.4352 0.3701 0.4000 0.9387

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.1.1+cu121
  • Datasets 2.14.5
  • Tokenizers 0.19.1