Hugging Face's logo

haryoaw
/
scenario-non-kd-scr-ner-full-xlmr_data-univner_full44

Token Classification
Transformers
Safetensors
xlm-roberta
Generated from Trainer
Inference Endpoints
Model card Files Community
scenario-non-kd-scr-ner-full-xlmr_data-univner_full44 / README.md
haryoaw's picture
haryoaw
Upload tokenizer
1178d59 verified
preview code
|
raw
history blame
No virus
11.3 kB
metadata
base_model: FacebookAI/xlm-roberta-base
library_name: transformers
license: mit
metrics:
  - precision
  - recall
  - f1
  - accuracy
tags:
  - generated_from_trainer
model-index:
  - name: scenario-non-kd-scr-ner-full-xlmr_data-univner_full44
    results: []

scenario-non-kd-scr-ner-full-xlmr_data-univner_full44

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3854
  • Precision: 0.5785
  • Recall: 0.5835
  • F1: 0.5810
  • Accuracy: 0.9605

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
0.338 0.2910 500 0.2757 0.375 0.1450 0.2091 0.9304
0.2636 0.5821 1000 0.2477 0.2864 0.2401 0.2612 0.9323
0.2294 0.8731 1500 0.2256 0.3407 0.2946 0.3160 0.9369
0.2017 1.1641 2000 0.2149 0.3468 0.3505 0.3486 0.9379
0.1872 1.4552 2500 0.2112 0.3721 0.3305 0.3501 0.9387
0.1753 1.7462 3000 0.2087 0.3904 0.3083 0.3445 0.9423
0.1623 2.0373 3500 0.1986 0.3818 0.3929 0.3873 0.9417
0.1324 2.3283 4000 0.1986 0.4314 0.4066 0.4186 0.9458
0.1307 2.6193 4500 0.1901 0.4309 0.4376 0.4342 0.9474
0.1199 2.9104 5000 0.1789 0.4454 0.4336 0.4394 0.9485
0.0995 3.2014 5500 0.1831 0.4761 0.4709 0.4735 0.9514
0.0895 3.4924 6000 0.1899 0.4646 0.5120 0.4872 0.9515
0.0827 3.7835 6500 0.1757 0.5053 0.5158 0.5105 0.9535
0.0761 4.0745 7000 0.2069 0.5092 0.4826 0.4956 0.9539
0.0616 4.3655 7500 0.1927 0.5246 0.5012 0.5127 0.9548
0.0609 4.6566 8000 0.1891 0.5251 0.4911 0.5076 0.9547
0.0577 4.9476 8500 0.1873 0.4978 0.5341 0.5153 0.9546
0.0451 5.2386 9000 0.2040 0.5239 0.5302 0.5270 0.9555
0.0419 5.5297 9500 0.2065 0.5249 0.5195 0.5222 0.9555
0.0433 5.8207 10000 0.2139 0.5171 0.5493 0.5327 0.9556
0.0392 6.1118 10500 0.2184 0.5140 0.5578 0.5350 0.9557
0.0308 6.4028 11000 0.2159 0.5110 0.5582 0.5335 0.9559
0.0312 6.6938 11500 0.2202 0.4900 0.5969 0.5382 0.9541
0.0296 6.9849 12000 0.2288 0.5260 0.5260 0.5260 0.9567
0.024 7.2759 12500 0.2368 0.5330 0.5667 0.5493 0.9572
0.0232 7.5669 13000 0.2438 0.5247 0.5399 0.5322 0.9565
0.0222 7.8580 13500 0.2483 0.5643 0.5266 0.5448 0.9573
0.0187 8.1490 14000 0.2476 0.5615 0.5333 0.5470 0.9567
0.0176 8.4400 14500 0.2473 0.5494 0.5445 0.5470 0.9571
0.0176 8.7311 15000 0.2452 0.5346 0.5715 0.5524 0.9570
0.0159 9.0221 15500 0.2794 0.5324 0.5569 0.5444 0.9577
0.0131 9.3132 16000 0.2672 0.5424 0.5917 0.5660 0.9582
0.0132 9.6042 16500 0.2716 0.5287 0.5566 0.5423 0.9572
0.0133 9.8952 17000 0.2668 0.5267 0.5669 0.5461 0.9569
0.0113 10.1863 17500 0.2779 0.5369 0.5770 0.5562 0.9578
0.0094 10.4773 18000 0.2717 0.5380 0.5881 0.5619 0.9578
0.0111 10.7683 18500 0.2861 0.5582 0.5465 0.5523 0.9587
0.0094 11.0594 19000 0.2803 0.5365 0.5833 0.5589 0.9582
0.008 11.3504 19500 0.2853 0.5262 0.5755 0.5498 0.9575
0.0077 11.6414 20000 0.2893 0.5366 0.5806 0.5577 0.9579
0.0083 11.9325 20500 0.2898 0.5415 0.5923 0.5657 0.9584
0.0067 12.2235 21000 0.3000 0.5635 0.5419 0.5525 0.9582
0.0066 12.5146 21500 0.3046 0.5574 0.5643 0.5608 0.9587
0.0065 12.8056 22000 0.3063 0.5495 0.5748 0.5619 0.9587
0.0062 13.0966 22500 0.3147 0.5619 0.5575 0.5597 0.9585
0.0056 13.3877 23000 0.3033 0.5440 0.5836 0.5631 0.9586
0.005 13.6787 23500 0.3083 0.5567 0.5741 0.5653 0.9585
0.0051 13.9697 24000 0.3201 0.5510 0.5891 0.5694 0.9592
0.0041 14.2608 24500 0.3265 0.5445 0.5687 0.5563 0.9586
0.0043 14.5518 25000 0.3202 0.5634 0.5641 0.5638 0.9586
0.0045 14.8428 25500 0.3200 0.5704 0.5677 0.5691 0.9597
0.0042 15.1339 26000 0.3285 0.5651 0.5770 0.5710 0.9595
0.0035 15.4249 26500 0.3259 0.5575 0.5846 0.5707 0.9590
0.0038 15.7159 27000 0.3300 0.5711 0.5685 0.5698 0.9595
0.0034 16.0070 27500 0.3244 0.5552 0.5771 0.5660 0.9586
0.003 16.2980 28000 0.3310 0.5683 0.5778 0.5730 0.9596
0.0031 16.5891 28500 0.3295 0.5629 0.5800 0.5713 0.9595
0.0029 16.8801 29000 0.3302 0.5396 0.5992 0.5679 0.9584
0.0028 17.1711 29500 0.3350 0.5519 0.5826 0.5668 0.9592
0.0029 17.4622 30000 0.3269 0.5360 0.6158 0.5732 0.9580
0.0023 17.7532 30500 0.3400 0.5731 0.5801 0.5766 0.9598
0.0022 18.0442 31000 0.3345 0.5716 0.5649 0.5682 0.9593
0.0022 18.3353 31500 0.3301 0.5589 0.5966 0.5771 0.9594
0.002 18.6263 32000 0.3406 0.5702 0.5774 0.5738 0.9596
0.0025 18.9173 32500 0.3422 0.5943 0.5380 0.5647 0.9595
0.0019 19.2084 33000 0.3476 0.5783 0.5728 0.5755 0.9600
0.0019 19.4994 33500 0.3449 0.5620 0.5910 0.5761 0.9596
0.0016 19.7905 34000 0.3518 0.5634 0.5926 0.5776 0.9595
0.0014 20.0815 34500 0.3522 0.5633 0.5820 0.5725 0.9598
0.0014 20.3725 35000 0.3486 0.5744 0.5739 0.5742 0.9598
0.0014 20.6636 35500 0.3513 0.5762 0.5737 0.5749 0.9596
0.0013 20.9546 36000 0.3608 0.5406 0.5851 0.5619 0.9586
0.0015 21.2456 36500 0.3548 0.5814 0.5696 0.5754 0.9602
0.0011 21.5367 37000 0.3552 0.5829 0.5700 0.5764 0.9600
0.0013 21.8277 37500 0.3565 0.5623 0.5797 0.5709 0.9593
0.0014 22.1187 38000 0.3608 0.5791 0.5693 0.5742 0.9603
0.001 22.4098 38500 0.3534 0.5706 0.5914 0.5808 0.9601
0.001 22.7008 39000 0.3639 0.5887 0.5719 0.5802 0.9603
0.0009 22.9919 39500 0.3650 0.5679 0.5898 0.5787 0.9597
0.0008 23.2829 40000 0.3676 0.5815 0.5744 0.5779 0.9602
0.0007 23.5739 40500 0.3738 0.5944 0.5680 0.5809 0.9606
0.001 23.8650 41000 0.3700 0.5804 0.5735 0.5769 0.9597
0.0009 24.1560 41500 0.3706 0.5774 0.5768 0.5771 0.9601
0.0008 24.4470 42000 0.3696 0.5838 0.5731 0.5784 0.9604
0.0007 24.7381 42500 0.3737 0.5656 0.5862 0.5757 0.9598
0.0007 25.0291 43000 0.3756 0.5617 0.5871 0.5741 0.9594
0.0006 25.3201 43500 0.3757 0.5668 0.5885 0.5774 0.9593
0.0004 25.6112 44000 0.3783 0.5865 0.5708 0.5785 0.9602
0.0006 25.9022 44500 0.3688 0.5724 0.5902 0.5812 0.9600
0.0004 26.1932 45000 0.3783 0.5851 0.5787 0.5819 0.9605
0.0004 26.4843 45500 0.3809 0.5773 0.5780 0.5776 0.9601
0.0004 26.7753 46000 0.3816 0.5823 0.5803 0.5813 0.9605
0.0004 27.0664 46500 0.3887 0.5762 0.5820 0.5791 0.9601
0.0003 27.3574 47000 0.3833 0.5834 0.5869 0.5852 0.9605
0.0004 27.6484 47500 0.3867 0.5791 0.5891 0.5840 0.9605
0.0004 27.9395 48000 0.3876 0.5814 0.5856 0.5835 0.9605
0.0003 28.2305 48500 0.3918 0.5810 0.5742 0.5776 0.9605
0.0003 28.5215 49000 0.3869 0.5804 0.5826 0.5815 0.9606
0.0003 28.8126 49500 0.3864 0.5760 0.5871 0.5815 0.9604
0.0004 29.1036 50000 0.3840 0.5732 0.5887 0.5808 0.9604
0.0003 29.3946 50500 0.3864 0.5833 0.5784 0.5808 0.9606
0.0002 29.6857 51000 0.3852 0.5781 0.5842 0.5811 0.9604
0.0002 29.9767 51500 0.3854 0.5785 0.5835 0.5810 0.9605

Framework versions

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