File size: 4,454 Bytes
f80157a ca0a4fd f80157a ca0a4fd f80157a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 |
---
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-non-kd-pre-ner-full-xlmr_data-univner_half55
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# scenario-non-kd-pre-ner-full-xlmr_data-univner_half55
This model is a fine-tuned version of [haryoaw/scenario-TCR-NER_data-univner_half](https://huggingface.co/haryoaw/scenario-TCR-NER_data-univner_half) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1450
- Precision: 0.8516
- Recall: 0.8357
- F1: 0.8436
- Accuracy: 0.9832
## 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: 55
- 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 |
|:-------------:|:-------:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
| 0.0106 | 0.5828 | 500 | 0.0804 | 0.8385 | 0.8489 | 0.8437 | 0.9838 |
| 0.0101 | 1.1655 | 1000 | 0.0863 | 0.8549 | 0.8432 | 0.8490 | 0.9838 |
| 0.009 | 1.7483 | 1500 | 0.0960 | 0.8282 | 0.8557 | 0.8418 | 0.9829 |
| 0.0077 | 2.3310 | 2000 | 0.1040 | 0.8320 | 0.8498 | 0.8408 | 0.9827 |
| 0.0077 | 2.9138 | 2500 | 0.0931 | 0.8461 | 0.8478 | 0.8469 | 0.9835 |
| 0.0054 | 3.4965 | 3000 | 0.0982 | 0.8482 | 0.8523 | 0.8502 | 0.9843 |
| 0.0061 | 4.0793 | 3500 | 0.1075 | 0.8450 | 0.8355 | 0.8402 | 0.9832 |
| 0.0053 | 4.6620 | 4000 | 0.1068 | 0.8409 | 0.8536 | 0.8472 | 0.9839 |
| 0.0047 | 5.2448 | 4500 | 0.1073 | 0.8372 | 0.8592 | 0.8480 | 0.9837 |
| 0.0047 | 5.8275 | 5000 | 0.1100 | 0.8451 | 0.8585 | 0.8517 | 0.9839 |
| 0.0034 | 6.4103 | 5500 | 0.1167 | 0.8350 | 0.8497 | 0.8422 | 0.9832 |
| 0.0034 | 6.9930 | 6000 | 0.1122 | 0.8405 | 0.8478 | 0.8441 | 0.9836 |
| 0.0035 | 7.5758 | 6500 | 0.1125 | 0.8424 | 0.8419 | 0.8421 | 0.9834 |
| 0.0032 | 8.1585 | 7000 | 0.1145 | 0.8454 | 0.8504 | 0.8479 | 0.9836 |
| 0.0035 | 8.7413 | 7500 | 0.1075 | 0.8499 | 0.8407 | 0.8453 | 0.9838 |
| 0.0027 | 9.3240 | 8000 | 0.1213 | 0.8493 | 0.8384 | 0.8438 | 0.9837 |
| 0.0031 | 9.9068 | 8500 | 0.1083 | 0.8551 | 0.8440 | 0.8495 | 0.9842 |
| 0.0027 | 10.4895 | 9000 | 0.1273 | 0.8329 | 0.8639 | 0.8482 | 0.9835 |
| 0.0024 | 11.0723 | 9500 | 0.1247 | 0.8478 | 0.8411 | 0.8444 | 0.9834 |
| 0.0021 | 11.6550 | 10000 | 0.1161 | 0.8487 | 0.8378 | 0.8432 | 0.9838 |
| 0.0019 | 12.2378 | 10500 | 0.1284 | 0.8316 | 0.8556 | 0.8434 | 0.9830 |
| 0.0021 | 12.8205 | 11000 | 0.1208 | 0.8492 | 0.8510 | 0.8501 | 0.9840 |
| 0.0015 | 13.4033 | 11500 | 0.1266 | 0.8374 | 0.8499 | 0.8436 | 0.9830 |
| 0.002 | 13.9860 | 12000 | 0.1236 | 0.8403 | 0.8530 | 0.8466 | 0.9832 |
| 0.0016 | 14.5688 | 12500 | 0.1313 | 0.8453 | 0.8409 | 0.8430 | 0.9833 |
| 0.0013 | 15.1515 | 13000 | 0.1362 | 0.8460 | 0.8482 | 0.8471 | 0.9835 |
| 0.0015 | 15.7343 | 13500 | 0.1246 | 0.8480 | 0.8511 | 0.8496 | 0.9840 |
| 0.0012 | 16.3170 | 14000 | 0.1335 | 0.8549 | 0.8423 | 0.8485 | 0.9837 |
| 0.0014 | 16.8998 | 14500 | 0.1265 | 0.8445 | 0.8433 | 0.8439 | 0.9833 |
| 0.0009 | 17.4825 | 15000 | 0.1450 | 0.8516 | 0.8357 | 0.8436 | 0.9832 |
### Framework versions
- Transformers 4.44.2
- Pytorch 2.1.1+cu121
- Datasets 2.14.5
- Tokenizers 0.19.1
|