Hugging Face's logo

InfoDeskInc
/
InfoDeskGenericNer

Safetensors
xlm-roberta
Generated from Trainer
Model card Files Community
InfoDeskGenericNer / README.md
vatsaldin's picture
vatsaldin
finetuned model with ontotext data
480dfe3 verified
preview code
|
raw
history blame
25.9 kB
metadata
license: mit
base_model: xlm-roberta-base
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: xlm-roberta-base-finetuned-generic_ner_ontonotes-ner-2024_08_14
    results: []

xlm-roberta-base-finetuned-generic_ner_ontonotes-ner-2024_08_14

This model is a fine-tuned version of xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0851
  • Precision: 0.8634
  • Recall: 0.8879
  • F1: 0.8755
  • Accuracy: 0.9783
  • O Precision: 0.9952
  • O Recall: 0.9917
  • O F1: 0.9934
  • B-cardinal Precision: 0.8585
  • B-cardinal Recall: 0.8994
  • B-cardinal F1: 0.8784
  • B-date Precision: 0.8627
  • B-date Recall: 0.8796
  • B-date F1: 0.8711
  • I-date Precision: 0.8742
  • I-date Recall: 0.9023
  • I-date F1: 0.8880
  • B-person Precision: 0.9204
  • B-person Recall: 0.9596
  • B-person F1: 0.9396
  • I-person Precision: 0.9452
  • I-person Recall: 0.9818
  • I-person F1: 0.9632
  • B-norp Precision: 0.8898
  • B-norp Recall: 0.9311
  • B-norp F1: 0.9100
  • B-gpe Precision: 0.9471
  • B-gpe Recall: 0.9395
  • B-gpe F1: 0.9433
  • I-gpe Precision: 0.9119
  • I-gpe Recall: 0.8846
  • I-gpe F1: 0.8980
  • B-law Precision: 0.5909
  • B-law Recall: 0.8667
  • B-law F1: 0.7027
  • I-law Precision: 0.5170
  • I-law Recall: 0.7982
  • I-law F1: 0.6276
  • B-org Precision: 0.9061
  • B-org Recall: 0.8716
  • B-org F1: 0.8885
  • I-org Precision: 0.9212
  • I-org Recall: 0.9075
  • I-org F1: 0.9143
  • B-percent Precision: 0.9321
  • B-percent Recall: 0.8996
  • B-percent F1: 0.9156
  • I-percent Precision: 0.8822
  • I-percent Recall: 0.9887
  • I-percent F1: 0.9324
  • B-ordinal Precision: 0.8356
  • B-ordinal Recall: 0.8356
  • B-ordinal F1: 0.8356
  • B-money Precision: 0.9051
  • B-money Recall: 0.9304
  • B-money F1: 0.9176
  • I-money Precision: 0.9372
  • I-money Recall: 0.9753
  • I-money F1: 0.9558
  • B-work Of Art Precision: 0.5354
  • B-work Of Art Recall: 0.6355
  • B-work Of Art F1: 0.5812
  • I-work Of Art Precision: 0.5849
  • I-work Of Art Recall: 0.6998
  • I-work Of Art F1: 0.6372
  • B-fac Precision: 0.4833
  • B-fac Recall: 0.6312
  • B-fac F1: 0.5474
  • B-time Precision: 0.7782
  • B-time Recall: 0.8299
  • B-time F1: 0.8032
  • I-cardinal Precision: 0.7683
  • I-cardinal Recall: 0.8892
  • I-cardinal F1: 0.8243
  • B-loc Precision: 0.8206
  • B-loc Recall: 0.7530
  • B-loc F1: 0.7854
  • B-quantity Precision: 0.8731
  • B-quantity Recall: 0.9
  • B-quantity F1: 0.8864
  • I-quantity Precision: 0.8889
  • I-quantity Recall: 0.9706
  • I-quantity F1: 0.9279
  • I-norp Precision: 0.6792
  • I-norp Recall: 0.5373
  • I-norp F1: 0.6000
  • I-loc Precision: 0.7721
  • I-loc Recall: 0.7692
  • I-loc F1: 0.7706
  • B-product Precision: 0.5447
  • B-product Recall: 0.6979
  • B-product F1: 0.6119
  • I-time Precision: 0.7694
  • I-time Recall: 0.8766
  • I-time F1: 0.8195
  • B-event Precision: 0.7308
  • B-event Recall: 0.5733
  • B-event F1: 0.6425
  • I-event Precision: 0.7951
  • I-event Recall: 0.6198
  • I-event F1: 0.6966
  • I-fac Precision: 0.6463
  • I-fac Recall: 0.6909
  • I-fac F1: 0.6678
  • B-language Precision: 0.8387
  • B-language Recall: 0.65
  • B-language F1: 0.7324
  • I-product Precision: 0.8480
  • I-product Recall: 0.8192
  • I-product F1: 0.8333
  • I-ordinal Precision: 1.0
  • I-ordinal Recall: 0.0
  • I-ordinal F1: 0.0
  • I-language Precision: 1.0
  • I-language Recall: 1.0
  • I-language F1: 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
  • lr_scheduler_warmup_steps: 500
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy O Precision O Recall O F1 B-cardinal Precision B-cardinal Recall B-cardinal F1 B-date Precision B-date Recall B-date F1 I-date Precision I-date Recall I-date F1 B-person Precision B-person Recall B-person F1 I-person Precision I-person Recall I-person F1 B-norp Precision B-norp Recall B-norp F1 B-gpe Precision B-gpe Recall B-gpe F1 I-gpe Precision I-gpe Recall I-gpe F1 B-law Precision B-law Recall B-law F1 I-law Precision I-law Recall I-law F1 B-org Precision B-org Recall B-org F1 I-org Precision I-org Recall I-org F1 B-percent Precision B-percent Recall B-percent F1 I-percent Precision I-percent Recall I-percent F1 B-ordinal Precision B-ordinal Recall B-ordinal F1 B-money Precision B-money Recall B-money F1 I-money Precision I-money Recall I-money F1 B-work Of Art Precision B-work Of Art Recall B-work Of Art F1 I-work Of Art Precision I-work Of Art Recall I-work Of Art F1 B-fac Precision B-fac Recall B-fac F1 B-time Precision B-time Recall B-time F1 I-cardinal Precision I-cardinal Recall I-cardinal F1 B-loc Precision B-loc Recall B-loc F1 B-quantity Precision B-quantity Recall B-quantity F1 I-quantity Precision I-quantity Recall I-quantity F1 I-norp Precision I-norp Recall I-norp F1 I-loc Precision I-loc Recall I-loc F1 B-product Precision B-product Recall B-product F1 I-time Precision I-time Recall I-time F1 B-event Precision B-event Recall B-event F1 I-event Precision I-event Recall I-event F1 I-fac Precision I-fac Recall I-fac F1 B-language Precision B-language Recall B-language F1 I-product Precision I-product Recall I-product F1 I-ordinal Precision I-ordinal Recall I-ordinal F1 I-language Precision I-language Recall I-language F1
0.43 0.3332 1248 0.1141 0.7842 0.8162 0.7999 0.9680 0.9940 0.9886 0.9913 0.8180 0.8605 0.8387 0.8343 0.8244 0.8294 0.8103 0.9023 0.8538 0.8776 0.9537 0.9141 0.9234 0.9693 0.9458 0.8485 0.9146 0.8803 0.9480 0.8583 0.9009 0.8588 0.7931 0.8246 1.0 0.0 0.0 0.3333 0.6754 0.4464 0.8220 0.8229 0.8224 0.8696 0.8734 0.8715 0.8902 0.9563 0.9221 0.9170 0.9170 0.9170 0.825 0.7911 0.8077 0.7884 0.8719 0.8280 0.9129 0.9593 0.9355 0.3367 0.1542 0.2115 0.4216 0.6862 0.5223 0.4031 0.325 0.3599 0.6897 0.4149 0.5181 0.6894 0.9125 0.7854 0.4729 0.6128 0.5339 0.8872 0.9077 0.8973 0.8442 0.9559 0.8966 0.75 0.0448 0.0845 0.5819 0.6374 0.6084 0.2260 0.6875 0.3402 0.7253 0.7437 0.7344 0.6857 0.2069 0.3179 0.5475 0.6832 0.6078 0.4843 0.6727 0.5632 1.0 0.025 0.0488 0.5455 0.1695 0.2586 1.0 0.0 0.0 1.0 1.0 1.0
0.105 0.6663 2496 0.0947 0.8324 0.8524 0.8423 0.9745 0.9920 0.9929 0.9924 0.8393 0.8623 0.8507 0.8746 0.8149 0.8437 0.8787 0.8613 0.8699 0.9319 0.9185 0.9252 0.9546 0.9673 0.9609 0.8904 0.9124 0.9012 0.8901 0.9615 0.9244 0.8396 0.9025 0.8699 0.6 0.4 0.48 0.4565 0.5526 0.5 0.8549 0.8488 0.8519 0.9132 0.8534 0.8823 0.9579 0.8952 0.9255 0.9203 0.9585 0.9390 0.8270 0.8185 0.8227 0.8616 0.9192 0.8895 0.9241 0.9738 0.9483 0.5193 0.5654 0.5414 0.5861 0.6682 0.6245 0.4261 0.4688 0.4464 0.7593 0.6805 0.7177 0.8645 0.8367 0.8504 0.8719 0.5396 0.6667 0.8769 0.8769 0.8769 0.8629 0.9485 0.9037 0.7179 0.4179 0.5283 0.8259 0.6081 0.7004 0.4889 0.6875 0.5714 0.74 0.8196 0.7778 0.7566 0.4957 0.5990 0.7438 0.6556 0.6969 0.6076 0.6364 0.6217 0.5246 0.8 0.6337 0.8489 0.6667 0.7468 1.0 0.0 0.0 1.0 1.0 1.0
0.0911 0.9995 3744 0.0930 0.8390 0.8678 0.8531 0.9746 0.9946 0.9902 0.9924 0.8656 0.8358 0.8505 0.8726 0.8452 0.8587 0.8527 0.9116 0.8812 0.8653 0.9684 0.9140 0.9111 0.9855 0.9468 0.8994 0.9176 0.9084 0.9277 0.9374 0.9325 0.8923 0.8696 0.8808 0.5667 0.7556 0.6476 0.4024 0.5789 0.4748 0.8818 0.8469 0.8640 0.8967 0.8967 0.8967 0.9327 0.9083 0.9204 0.8680 0.9925 0.9261 0.8854 0.7671 0.8220 0.8770 0.9331 0.9042 0.9266 0.9724 0.9489 0.4681 0.6168 0.5323 0.5297 0.6637 0.5892 0.5475 0.6125 0.5782 0.72 0.7469 0.7332 0.7570 0.8717 0.8103 0.8699 0.6524 0.7456 0.8188 0.8692 0.8433 0.8248 0.9522 0.8840 0.6364 0.5224 0.5738 0.8806 0.6484 0.7468 0.5138 0.5833 0.5463 0.7733 0.8418 0.8061 0.7319 0.4353 0.5459 0.7147 0.6556 0.6839 0.6151 0.6218 0.6184 0.7429 0.65 0.6933 0.7669 0.7062 0.7353 1.0 0.0 0.0 1.0 1.0 1.0
0.0687 1.3326 4992 0.0859 0.8593 0.8739 0.8665 0.9773 0.9938 0.9923 0.9930 0.8537 0.8959 0.8742 0.8901 0.8505 0.8699 0.8793 0.8914 0.8853 0.9347 0.9473 0.9410 0.9343 0.9822 0.9577 0.9099 0.9154 0.9126 0.9510 0.9254 0.9380 0.8698 0.8816 0.8757 0.6383 0.6667 0.6522 0.6562 0.5526 0.6 0.8763 0.8787 0.8775 0.8960 0.9126 0.9042 0.9298 0.9258 0.9278 0.8931 0.9774 0.9333 0.8322 0.8151 0.8235 0.9126 0.9304 0.9214 0.9372 0.9753 0.9558 0.5642 0.5748 0.5694 0.5895 0.6614 0.6234 0.4837 0.5563 0.5174 0.7592 0.7718 0.7654 0.7927 0.8805 0.8343 0.7432 0.75 0.7466 0.8551 0.9077 0.8806 0.8961 0.9191 0.9074 0.75 0.4925 0.5946 0.7807 0.7692 0.7749 0.5285 0.6771 0.5936 0.7326 0.8671 0.7942 0.7314 0.5517 0.6290 0.7697 0.6722 0.7176 0.6778 0.6655 0.6716 0.7368 0.7 0.7179 0.8531 0.6893 0.7625 1.0 0.0 0.0 1.0 1.0 1.0
0.0638 1.6658 6240 0.0855 0.8512 0.8825 0.8665 0.9769 0.9947 0.9912 0.9930 0.8476 0.8985 0.8723 0.8660 0.8737 0.8698 0.8713 0.8919 0.8815 0.9208 0.9577 0.9389 0.9413 0.9786 0.9596 0.9022 0.9258 0.9139 0.9359 0.9535 0.9446 0.9280 0.8891 0.9081 0.5818 0.7111 0.64 0.5780 0.5526 0.5650 0.8955 0.8685 0.8818 0.9273 0.8887 0.9076 0.9185 0.9345 0.9264 0.9231 0.9509 0.9368 0.8530 0.8151 0.8336 0.9187 0.9443 0.9313 0.9579 0.9593 0.9586 0.4462 0.6776 0.5380 0.4534 0.7472 0.5644 0.4554 0.6375 0.5312 0.7578 0.8050 0.7807 0.7677 0.8863 0.8227 0.8517 0.6829 0.7580 0.8369 0.9077 0.8708 0.8854 0.9375 0.9107 0.8919 0.4925 0.6346 0.8918 0.6337 0.7409 0.5349 0.7188 0.6133 0.7778 0.8418 0.8085 0.7590 0.5431 0.6332 0.7812 0.6198 0.6912 0.5287 0.6691 0.5907 0.8214 0.575 0.6765 0.8447 0.7684 0.8047 1.0 0.0 0.0 1.0 1.0 1.0
0.059 1.9989 7488 0.0828 0.8576 0.8837 0.8705 0.9778 0.9945 0.9919 0.9932 0.8352 0.9038 0.8682 0.8651 0.8749 0.8699 0.8784 0.8875 0.8829 0.9139 0.9589 0.9359 0.9361 0.9826 0.9588 0.8975 0.9311 0.9140 0.9323 0.9481 0.9401 0.8806 0.9175 0.8987 0.6 0.6667 0.6316 0.576 0.6316 0.6025 0.8991 0.8673 0.8829 0.9162 0.9005 0.9083 0.9330 0.9127 0.9227 0.8966 0.9811 0.9369 0.7907 0.8151 0.8027 0.8939 0.9387 0.9158 0.9448 0.9709 0.9577 0.5774 0.6449 0.6093 0.6834 0.6772 0.6803 0.5024 0.6438 0.5644 0.7870 0.7510 0.7686 0.7655 0.8659 0.8126 0.8710 0.6585 0.75 0.8992 0.8923 0.8958 0.8969 0.9596 0.9272 0.8571 0.5373 0.6606 0.8517 0.6520 0.7386 0.5610 0.7188 0.6301 0.8185 0.8133 0.8159 0.7396 0.6121 0.6698 0.8033 0.6749 0.7335 0.5574 0.7418 0.6365 0.9259 0.625 0.7463 0.8333 0.7627 0.7965 1.0 0.0 0.0 1.0 1.0 1.0
0.0469 2.3321 8736 0.0843 0.8674 0.8854 0.8763 0.9785 0.9944 0.9921 0.9932 0.8763 0.8817 0.8790 0.8636 0.8826 0.8730 0.8583 0.9116 0.8842 0.9310 0.9556 0.9432 0.9548 0.9818 0.9681 0.8977 0.9266 0.9119 0.9468 0.9398 0.9433 0.9367 0.8876 0.9115 0.6 0.8 0.6857 0.6083 0.6404 0.6239 0.9025 0.8724 0.8872 0.9201 0.9043 0.9121 0.9361 0.8952 0.9152 0.8680 0.9925 0.9261 0.8339 0.8253 0.8296 0.9066 0.9192 0.9129 0.9385 0.9767 0.9573 0.5744 0.6495 0.6096 0.6268 0.6975 0.6603 0.4928 0.6375 0.5559 0.8 0.7801 0.7899 0.7943 0.9009 0.8443 0.7980 0.7226 0.7584 0.8992 0.8923 0.8958 0.8859 0.9706 0.9263 0.8182 0.5373 0.6486 0.8016 0.7399 0.7695 0.5185 0.7292 0.6061 0.7604 0.8639 0.8089 0.7273 0.6207 0.6698 0.7753 0.6749 0.7216 0.6168 0.72 0.6644 0.9259 0.625 0.7463 0.8114 0.8023 0.8068 1.0 0.0 0.0 1.0 1.0 1.0
0.0439 2.6652 9984 0.0868 0.8635 0.8846 0.8739 0.9781 0.9953 0.9914 0.9934 0.8517 0.8923 0.8716 0.8660 0.8778 0.8719 0.8604 0.9097 0.8844 0.9214 0.9623 0.9414 0.9377 0.9842 0.9604 0.8797 0.9258 0.9022 0.9402 0.9323 0.9362 0.8920 0.9040 0.8980 0.6066 0.8222 0.6981 0.4802 0.7456 0.5842 0.9044 0.8693 0.8865 0.9143 0.9122 0.9133 0.9321 0.8996 0.9156 0.8763 0.9887 0.9291 0.8385 0.8356 0.8370 0.9046 0.9248 0.9146 0.9321 0.9782 0.9546 0.6207 0.5888 0.6043 0.6463 0.6930 0.6688 0.5575 0.6062 0.5808 0.7787 0.8174 0.7976 0.7440 0.9067 0.8173 0.7508 0.7530 0.7519 0.8712 0.8846 0.8779 0.88 0.9706 0.9231 0.6792 0.5373 0.6000 0.7948 0.7802 0.7874 0.5378 0.6667 0.5953 0.7744 0.8797 0.8237 0.7366 0.5905 0.6555 0.7912 0.6474 0.7121 0.6809 0.6982 0.6894 0.6744 0.725 0.6988 0.7967 0.8192 0.8078 1.0 0.0 0.0 1.0 1.0 1.0
0.0435 2.9984 11232 0.0851 0.8634 0.8879 0.8755 0.9783 0.9952 0.9917 0.9934 0.8585 0.8994 0.8784 0.8627 0.8796 0.8711 0.8742 0.9023 0.8880 0.9204 0.9596 0.9396 0.9452 0.9818 0.9632 0.8898 0.9311 0.9100 0.9471 0.9395 0.9433 0.9119 0.8846 0.8980 0.5909 0.8667 0.7027 0.5170 0.7982 0.6276 0.9061 0.8716 0.8885 0.9212 0.9075 0.9143 0.9321 0.8996 0.9156 0.8822 0.9887 0.9324 0.8356 0.8356 0.8356 0.9051 0.9304 0.9176 0.9372 0.9753 0.9558 0.5354 0.6355 0.5812 0.5849 0.6998 0.6372 0.4833 0.6312 0.5474 0.7782 0.8299 0.8032 0.7683 0.8892 0.8243 0.8206 0.7530 0.7854 0.8731 0.9 0.8864 0.8889 0.9706 0.9279 0.6792 0.5373 0.6000 0.7721 0.7692 0.7706 0.5447 0.6979 0.6119 0.7694 0.8766 0.8195 0.7308 0.5733 0.6425 0.7951 0.6198 0.6966 0.6463 0.6909 0.6678 0.8387 0.65 0.7324 0.8480 0.8192 0.8333 1.0 0.0 0.0 1.0 1.0 1.0

Framework versions

  • Transformers 4.42.4
  • Pytorch 2.3.1+cu121
  • Datasets 2.21.0
  • Tokenizers 0.19.1