Hugging Face's logo

haryoaw
/
scenario-non-kd-pre-ner-full-mdeberta_data-univner_half66

Token Classification
Transformers
Safetensors
deberta-v2
Generated from Trainer
Inference Endpoints
Model card Files Community
scenario-non-kd-pre-ner-full-mdeberta_data-univner_half66 / README.md
haryoaw's picture
haryoaw
Upload tokenizer
1b847e7 verified
preview code
|
raw
history blame contribute delete
No virus
5.6 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-non-kd-pre-ner-full-mdeberta_data-univner_half66
    results: []

scenario-non-kd-pre-ner-full-mdeberta_data-univner_half66

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: 0.1403
  • Precision: 0.8613
  • Recall: 0.8638
  • F1: 0.8626
  • Accuracy: 0.9848

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: 66
  • 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
0.0064 0.5828 500 0.0946 0.8529 0.8691 0.8609 0.9847
0.0076 1.1655 1000 0.0939 0.8503 0.8562 0.8532 0.9842
0.0059 1.7483 1500 0.0980 0.8583 0.8515 0.8549 0.9840
0.005 2.3310 2000 0.1052 0.8469 0.8618 0.8543 0.9840
0.0054 2.9138 2500 0.1025 0.8389 0.8699 0.8541 0.9841
0.0045 3.4965 3000 0.1032 0.8371 0.8696 0.8530 0.9836
0.0042 4.0793 3500 0.1088 0.8459 0.8642 0.8550 0.9840
0.0032 4.6620 4000 0.1182 0.8301 0.8691 0.8492 0.9828
0.0035 5.2448 4500 0.1164 0.8486 0.8611 0.8548 0.9841
0.0031 5.8275 5000 0.1190 0.8352 0.8602 0.8475 0.9836
0.0029 6.4103 5500 0.1197 0.8516 0.8694 0.8604 0.9843
0.0029 6.9930 6000 0.1177 0.8282 0.8674 0.8474 0.9833
0.0024 7.5758 6500 0.1219 0.8396 0.8680 0.8536 0.9845
0.0031 8.1585 7000 0.1160 0.8566 0.8559 0.8562 0.9846
0.002 8.7413 7500 0.1222 0.8385 0.8624 0.8503 0.9834
0.0021 9.3240 8000 0.1217 0.8522 0.8667 0.8594 0.9847
0.0019 9.9068 8500 0.1333 0.8222 0.8699 0.8453 0.9835
0.002 10.4895 9000 0.1210 0.8475 0.8665 0.8569 0.9845
0.0017 11.0723 9500 0.1192 0.8571 0.8642 0.8606 0.9849
0.0013 11.6550 10000 0.1329 0.8524 0.8716 0.8619 0.9848
0.0016 12.2378 10500 0.1337 0.8493 0.8700 0.8595 0.9844
0.0014 12.8205 11000 0.1245 0.8635 0.8707 0.8671 0.9854
0.0014 13.4033 11500 0.1299 0.8611 0.8595 0.8603 0.9849
0.0012 13.9860 12000 0.1229 0.8545 0.8657 0.8600 0.9848
0.0011 14.5688 12500 0.1258 0.8585 0.8631 0.8608 0.9849
0.0008 15.1515 13000 0.1377 0.8558 0.8658 0.8608 0.9847
0.001 15.7343 13500 0.1328 0.8576 0.8611 0.8593 0.9846
0.0008 16.3170 14000 0.1331 0.8596 0.8660 0.8628 0.9850
0.0008 16.8998 14500 0.1292 0.8549 0.8694 0.8621 0.9849
0.0008 17.4825 15000 0.1388 0.8496 0.8699 0.8596 0.9846
0.0008 18.0653 15500 0.1364 0.8577 0.8629 0.8603 0.9848
0.0005 18.6480 16000 0.1419 0.8627 0.8645 0.8636 0.9848
0.0007 19.2308 16500 0.1414 0.8569 0.8709 0.8638 0.9850
0.0005 19.8135 17000 0.1369 0.8513 0.8700 0.8606 0.9848
0.0004 20.3963 17500 0.1419 0.8580 0.8658 0.8619 0.9849
0.0004 20.9790 18000 0.1452 0.8598 0.8700 0.8649 0.9849
0.0005 21.5618 18500 0.1417 0.8540 0.8673 0.8606 0.9842
0.0003 22.1445 19000 0.1419 0.8667 0.8611 0.8639 0.9848
0.0003 22.7273 19500 0.1500 0.8588 0.8632 0.8610 0.9845
0.0004 23.3100 20000 0.1470 0.8557 0.8717 0.8636 0.9846
0.0004 23.8928 20500 0.1387 0.8652 0.8671 0.8662 0.9852
0.0002 24.4755 21000 0.1403 0.8613 0.8638 0.8626 0.9848

Framework versions

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