robert_bilstm_mega_res-ner-resume-ner

This model is a fine-tuned version of hfl/chinese-roberta-wwm-ext-large on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2496
  • Precision: 0.9534
  • Recall: 0.9609
  • F1: 0.9572
  • Accuracy: 0.9756

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: 5e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • num_epochs: 100

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
0.0932 1.0 134 0.0917 0.9224 0.9496 0.9358 0.9734
0.0356 2.0 268 0.0776 0.9343 0.9635 0.9487 0.9777
0.0363 3.0 402 0.1026 0.9309 0.9479 0.9393 0.9727
0.0278 4.0 536 0.1041 0.9408 0.9531 0.9469 0.9750
0.0131 5.0 670 0.1194 0.9411 0.9574 0.9492 0.9746
0.0071 6.0 804 0.1491 0.9593 0.9618 0.9605 0.9768
0.0042 7.0 938 0.1474 0.9567 0.9609 0.9588 0.9749
0.0105 8.0 1072 0.1318 0.9527 0.9635 0.9581 0.9768
0.0039 9.0 1206 0.1616 0.9374 0.9496 0.9435 0.9724
0.0023 10.0 1340 0.1547 0.9452 0.9592 0.9521 0.9755
0.0026 11.0 1474 0.1585 0.9447 0.9505 0.9476 0.9746
0.0046 12.0 1608 0.1492 0.9411 0.9574 0.9492 0.9748
0.0051 13.0 1742 0.1440 0.9519 0.9626 0.9572 0.9755
0.0037 14.0 1876 0.1651 0.9516 0.9574 0.9545 0.9743
0.0006 15.0 2010 0.1641 0.9517 0.9583 0.9550 0.9755
0.0003 16.0 2144 0.1957 0.9536 0.9635 0.9585 0.9734
0.0002 17.0 2278 0.1896 0.9469 0.9600 0.9534 0.9734
0.0016 18.0 2412 0.1877 0.9419 0.9574 0.9496 0.9731
0.0001 19.0 2546 0.2117 0.9481 0.9531 0.9506 0.9722
0.0018 20.0 2680 0.2165 0.9540 0.9548 0.9544 0.9743
0.0011 21.0 2814 0.2082 0.9534 0.9592 0.9563 0.9719
0.0006 22.0 2948 0.2052 0.9567 0.9592 0.9579 0.9730
0.0001 23.0 3082 0.2016 0.9551 0.9600 0.9575 0.9746
0.0019 24.0 3216 0.1969 0.9549 0.9557 0.9553 0.9746
0.001 25.0 3350 0.1927 0.9516 0.9557 0.9536 0.9742
0.0002 26.0 3484 0.1767 0.9601 0.9609 0.9605 0.9752
0.0003 27.0 3618 0.1936 0.9433 0.9540 0.9486 0.9735
0.0011 28.0 3752 0.1928 0.9425 0.9548 0.9486 0.9734
0.0001 29.0 3886 0.2076 0.9456 0.9522 0.9489 0.9739
0.0052 30.0 4020 0.2040 0.9418 0.9557 0.9487 0.9748
0.0013 31.0 4154 0.1716 0.9552 0.9635 0.9593 0.9749
0.0004 32.0 4288 0.1762 0.9475 0.9566 0.9520 0.9755
0.0004 33.0 4422 0.1479 0.9450 0.9557 0.9503 0.9768
0.0001 34.0 4556 0.1668 0.9469 0.9609 0.9539 0.9766
0.0019 35.0 4690 0.1884 0.9508 0.9574 0.9541 0.9766
0.0001 36.0 4824 0.1873 0.9451 0.9566 0.9508 0.9748
0.0012 37.0 4958 0.1807 0.9553 0.9652 0.9602 0.9763
0.0071 38.0 5092 0.1993 0.9558 0.9592 0.9575 0.9760
0.0 39.0 5226 0.1821 0.9483 0.9557 0.9520 0.9751
0.0008 40.0 5360 0.1968 0.9576 0.9626 0.9601 0.9757
0.0011 41.0 5494 0.1930 0.9493 0.9600 0.9546 0.9751
0.0016 42.0 5628 0.1974 0.9476 0.9592 0.9534 0.9753
0.0 43.0 5762 0.1935 0.9477 0.9600 0.9538 0.9759
0.0015 44.0 5896 0.1970 0.9525 0.9583 0.9554 0.9757
0.0008 45.0 6030 0.2079 0.9542 0.9592 0.9567 0.9775
0.0003 46.0 6164 0.1899 0.9435 0.9583 0.9509 0.9746
0.0047 47.0 6298 0.1822 0.9478 0.9626 0.9552 0.9748
0.0019 48.0 6432 0.1990 0.9543 0.9609 0.9576 0.9758
0.001 49.0 6566 0.2001 0.9518 0.9609 0.9563 0.9764
0.0 50.0 6700 0.1961 0.9493 0.9600 0.9546 0.9758
0.0 51.0 6834 0.2083 0.9551 0.9618 0.9584 0.9765
0.0 52.0 6968 0.2072 0.9558 0.9583 0.9570 0.9762
0.0 53.0 7102 0.2112 0.9516 0.9574 0.9545 0.9764
0.0001 54.0 7236 0.2160 0.9560 0.9618 0.9589 0.9774
0.0 55.0 7370 0.2166 0.9560 0.9618 0.9589 0.9771
0.0 56.0 7504 0.2188 0.9560 0.9618 0.9589 0.9770
0.0 57.0 7638 0.2208 0.9568 0.9618 0.9593 0.9771
0.0007 58.0 7772 0.2301 0.9559 0.9600 0.9580 0.9764
0.0 59.0 7906 0.2317 0.9551 0.9600 0.9575 0.9764
0.0 60.0 8040 0.2326 0.9559 0.9600 0.9580 0.9759
0.0 61.0 8174 0.2349 0.9583 0.9592 0.9587 0.9761
0.0 62.0 8308 0.2378 0.9550 0.9592 0.9571 0.9768
0.0 63.0 8442 0.2406 0.9526 0.9600 0.9563 0.9767
0.0 64.0 8576 0.2367 0.9584 0.9600 0.9592 0.9771
0.0009 65.0 8710 0.2329 0.9584 0.9600 0.9592 0.9772
0.0 66.0 8844 0.2376 0.9524 0.9566 0.9545 0.9761
0.0009 67.0 8978 0.2381 0.9509 0.9583 0.9546 0.9759
0.0 68.0 9112 0.2236 0.9585 0.9644 0.9615 0.9778
0.0 69.0 9246 0.2194 0.9619 0.9644 0.9631 0.9769
0.0 70.0 9380 0.2192 0.9595 0.9670 0.9632 0.9775
0.0 71.0 9514 0.2124 0.9562 0.9679 0.9620 0.9777
0.0007 72.0 9648 0.2139 0.9587 0.9679 0.9633 0.9777
0.0 73.0 9782 0.2266 0.9552 0.9644 0.9598 0.9764
0.0 74.0 9916 0.2275 0.9593 0.9635 0.9614 0.9765
0.0 75.0 10050 0.2239 0.9577 0.9644 0.9610 0.9774
0.0 76.0 10184 0.2212 0.9578 0.9652 0.9615 0.9760
0.0 77.0 10318 0.2208 0.9560 0.9635 0.9598 0.9758
0.0 78.0 10452 0.2259 0.9577 0.9635 0.9606 0.9757
0.0 79.0 10586 0.2279 0.9593 0.9635 0.9614 0.9756
0.0 80.0 10720 0.2294 0.9577 0.9635 0.9606 0.9756
0.0 81.0 10854 0.2380 0.9492 0.9574 0.9533 0.9766
0.0 82.0 10988 0.2405 0.9577 0.9644 0.9610 0.9759
0.0 83.0 11122 0.2442 0.9511 0.9635 0.9573 0.9751
0.0003 84.0 11256 0.2371 0.9552 0.9626 0.9589 0.9749
0.0 85.0 11390 0.2499 0.9559 0.9609 0.9584 0.9750
0.0 86.0 11524 0.2516 0.9601 0.9626 0.9614 0.9752
0.0 87.0 11658 0.2519 0.9593 0.9626 0.9610 0.9752
0.0008 88.0 11792 0.2516 0.9576 0.9618 0.9597 0.9752
0.0007 89.0 11926 0.2517 0.9593 0.9626 0.9610 0.9753
0.0 90.0 12060 0.2477 0.9559 0.9609 0.9584 0.9754
0.0 91.0 12194 0.2477 0.9559 0.9609 0.9584 0.9754
0.0 92.0 12328 0.2479 0.9543 0.9609 0.9576 0.9755
0.0007 93.0 12462 0.2482 0.9543 0.9609 0.9576 0.9754
0.0 94.0 12596 0.2487 0.9534 0.9609 0.9572 0.9755
0.0 95.0 12730 0.2490 0.9534 0.9609 0.9572 0.9755
0.0 96.0 12864 0.2495 0.9534 0.9609 0.9572 0.9755
0.0 97.0 12998 0.2489 0.9534 0.9609 0.9572 0.9755
0.0 98.0 13132 0.2495 0.9534 0.9609 0.9572 0.9756
0.0006 99.0 13266 0.2496 0.9534 0.9609 0.9572 0.9756
0.0 100.0 13400 0.2496 0.9534 0.9609 0.9572 0.9756

Framework versions

  • Transformers 4.46.2
  • Pytorch 2.4.1+cu124
  • Datasets 3.1.0
  • Tokenizers 0.20.3
Downloads last month
17
Safetensors
Model size
324M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for PassbyGrocer/robert_bilstm_mega_res-ner-resume-ner

Finetuned
(6)
this model