Hugging Face's logo

thangvip
/
bert-30M-uncased-classification-CMC-fqa-new-100e

Text Classification
Transformers
TensorBoard
Safetensors
bert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
bert-30M-uncased-classification-CMC-fqa-new-100e / README.md
thangvip's picture
thangvip
End of training
a38f792 verified
preview code
|
raw
history blame contribute delete
7.58 kB
metadata
license: apache-2.0
base_model: vietgpt/bert-30M-uncased
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: bert-30M-uncased-classification-CMC-fqa-new-100e
    results: []

bert-30M-uncased-classification-CMC-fqa-new-100e

This model is a fine-tuned version of vietgpt/bert-30M-uncased on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0430
  • Accuracy: 1.0

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: 2e-05
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 100

Training results

Training Loss Epoch Step Validation Loss Accuracy
No log 1.0 20 3.4282 0.0323
No log 2.0 40 3.4133 0.0323
No log 3.0 60 3.4014 0.0645
No log 4.0 80 3.3872 0.1613
No log 5.0 100 3.3703 0.2581
No log 6.0 120 3.3503 0.2903
No log 7.0 140 3.3193 0.5161
No log 8.0 160 3.2707 0.5484
No log 9.0 180 3.1889 0.5484
No log 10.0 200 3.0520 0.4839
No log 11.0 220 2.8649 0.3548
No log 12.0 240 2.6818 0.4839
No log 13.0 260 2.5264 0.5161
No log 14.0 280 2.3816 0.5806
No log 15.0 300 2.2296 0.7419
No log 16.0 320 2.0829 0.7097
No log 17.0 340 1.9155 0.8387
No log 18.0 360 1.7785 0.8065
No log 19.0 380 1.6237 0.8710
No log 20.0 400 1.4944 0.9355
No log 21.0 420 1.3666 0.9355
No log 22.0 440 1.2578 0.9355
No log 23.0 460 1.1628 0.9677
No log 24.0 480 1.0483 0.9355
2.5506 25.0 500 0.9724 0.9677
2.5506 26.0 520 0.9024 0.9355
2.5506 27.0 540 0.8159 0.9677
2.5506 28.0 560 0.7614 0.9355
2.5506 29.0 580 0.6883 0.9677
2.5506 30.0 600 0.6365 0.9677
2.5506 31.0 620 0.5864 0.9677
2.5506 32.0 640 0.5423 1.0
2.5506 33.0 660 0.4985 1.0
2.5506 34.0 680 0.4666 1.0
2.5506 35.0 700 0.4262 1.0
2.5506 36.0 720 0.3942 1.0
2.5506 37.0 740 0.3629 1.0
2.5506 38.0 760 0.3336 1.0
2.5506 39.0 780 0.3079 1.0
2.5506 40.0 800 0.2836 1.0
2.5506 41.0 820 0.2579 1.0
2.5506 42.0 840 0.2381 1.0
2.5506 43.0 860 0.2167 1.0
2.5506 44.0 880 0.2053 1.0
2.5506 45.0 900 0.1846 1.0
2.5506 46.0 920 0.1720 1.0
2.5506 47.0 940 0.1579 1.0
2.5506 48.0 960 0.1452 1.0
2.5506 49.0 980 0.1366 1.0
0.4668 50.0 1000 0.1262 1.0
0.4668 51.0 1020 0.1165 1.0
0.4668 52.0 1040 0.1098 1.0
0.4668 53.0 1060 0.1042 1.0
0.4668 54.0 1080 0.1022 1.0
0.4668 55.0 1100 0.0956 1.0
0.4668 56.0 1120 0.0916 1.0
0.4668 57.0 1140 0.0875 1.0
0.4668 58.0 1160 0.0819 1.0
0.4668 59.0 1180 0.0785 1.0
0.4668 60.0 1200 0.0761 1.0
0.4668 61.0 1220 0.0740 1.0
0.4668 62.0 1240 0.0709 1.0
0.4668 63.0 1260 0.0692 1.0
0.4668 64.0 1280 0.0672 1.0
0.4668 65.0 1300 0.0654 1.0
0.4668 66.0 1320 0.0653 1.0
0.4668 67.0 1340 0.0628 1.0
0.4668 68.0 1360 0.0618 1.0
0.4668 69.0 1380 0.0608 1.0
0.4668 70.0 1400 0.0593 1.0
0.4668 71.0 1420 0.0582 1.0
0.4668 72.0 1440 0.0579 1.0
0.4668 73.0 1460 0.0558 1.0
0.4668 74.0 1480 0.0540 1.0
0.0861 75.0 1500 0.0528 1.0
0.0861 76.0 1520 0.0522 1.0
0.0861 77.0 1540 0.0516 1.0
0.0861 78.0 1560 0.0502 1.0
0.0861 79.0 1580 0.0494 1.0
0.0861 80.0 1600 0.0490 1.0
0.0861 81.0 1620 0.0484 1.0
0.0861 82.0 1640 0.0480 1.0
0.0861 83.0 1660 0.0479 1.0
0.0861 84.0 1680 0.0476 1.0
0.0861 85.0 1700 0.0470 1.0
0.0861 86.0 1720 0.0465 1.0
0.0861 87.0 1740 0.0455 1.0
0.0861 88.0 1760 0.0449 1.0
0.0861 89.0 1780 0.0444 1.0
0.0861 90.0 1800 0.0444 1.0
0.0861 91.0 1820 0.0445 1.0
0.0861 92.0 1840 0.0441 1.0
0.0861 93.0 1860 0.0439 1.0
0.0861 94.0 1880 0.0438 1.0
0.0861 95.0 1900 0.0436 1.0
0.0861 96.0 1920 0.0434 1.0
0.0861 97.0 1940 0.0433 1.0
0.0861 98.0 1960 0.0431 1.0
0.0861 99.0 1980 0.0430 1.0
0.0493 100.0 2000 0.0430 1.0

Framework versions

  • Transformers 4.37.1
  • Pytorch 2.1.0+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.1