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-pre-ner-full-xlmr_data-univner_half66
results: []
scenario-non-kd-pre-ner-full-xlmr_data-univner_half66
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: 0.1641
- Precision: 0.7981
- Recall: 0.8067
- F1: 0.8024
- Accuracy: 0.9795
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 |
---|---|---|---|---|---|---|---|
0.1478 | 0.5828 | 500 | 0.0839 | 0.6901 | 0.7785 | 0.7317 | 0.9723 |
0.0711 | 1.1655 | 1000 | 0.0786 | 0.7355 | 0.7997 | 0.7663 | 0.9768 |
0.0512 | 1.7483 | 1500 | 0.0727 | 0.7800 | 0.7846 | 0.7823 | 0.9784 |
0.037 | 2.3310 | 2000 | 0.0811 | 0.7803 | 0.7855 | 0.7829 | 0.9777 |
0.0312 | 2.9138 | 2500 | 0.0812 | 0.7647 | 0.8139 | 0.7885 | 0.9781 |
0.0216 | 3.4965 | 3000 | 0.0914 | 0.7814 | 0.8009 | 0.7910 | 0.9783 |
0.0203 | 4.0793 | 3500 | 0.0944 | 0.7849 | 0.8103 | 0.7974 | 0.9789 |
0.0146 | 4.6620 | 4000 | 0.1050 | 0.7516 | 0.8172 | 0.7830 | 0.9770 |
0.0129 | 5.2448 | 4500 | 0.1067 | 0.7649 | 0.8064 | 0.7851 | 0.9783 |
0.0112 | 5.8275 | 5000 | 0.1087 | 0.7743 | 0.8061 | 0.7898 | 0.9781 |
0.0101 | 6.4103 | 5500 | 0.1124 | 0.7820 | 0.7990 | 0.7904 | 0.9783 |
0.0097 | 6.9930 | 6000 | 0.1137 | 0.7716 | 0.7907 | 0.7810 | 0.9783 |
0.0072 | 7.5758 | 6500 | 0.1171 | 0.7946 | 0.7908 | 0.7927 | 0.9785 |
0.007 | 8.1585 | 7000 | 0.1187 | 0.7866 | 0.8059 | 0.7962 | 0.9787 |
0.0057 | 8.7413 | 7500 | 0.1204 | 0.7930 | 0.8002 | 0.7966 | 0.9787 |
0.0052 | 9.3240 | 8000 | 0.1264 | 0.7876 | 0.8077 | 0.7975 | 0.9791 |
0.0047 | 9.9068 | 8500 | 0.1277 | 0.7887 | 0.8041 | 0.7963 | 0.9786 |
0.0047 | 10.4895 | 9000 | 0.1324 | 0.7687 | 0.8133 | 0.7904 | 0.9781 |
0.0041 | 11.0723 | 9500 | 0.1233 | 0.7965 | 0.7987 | 0.7976 | 0.9790 |
0.0037 | 11.6550 | 10000 | 0.1298 | 0.7902 | 0.7961 | 0.7932 | 0.9788 |
0.0037 | 12.2378 | 10500 | 0.1327 | 0.7757 | 0.8143 | 0.7945 | 0.9787 |
0.0031 | 12.8205 | 11000 | 0.1353 | 0.7816 | 0.8054 | 0.7933 | 0.9788 |
0.003 | 13.4033 | 11500 | 0.1378 | 0.7987 | 0.7895 | 0.7941 | 0.9788 |
0.0027 | 13.9860 | 12000 | 0.1414 | 0.7928 | 0.7931 | 0.7929 | 0.9788 |
0.0024 | 14.5688 | 12500 | 0.1398 | 0.7851 | 0.8044 | 0.7946 | 0.9787 |
0.0024 | 15.1515 | 13000 | 0.1396 | 0.7792 | 0.8181 | 0.7981 | 0.9792 |
0.0017 | 15.7343 | 13500 | 0.1438 | 0.7934 | 0.8025 | 0.7979 | 0.9791 |
0.0021 | 16.3170 | 14000 | 0.1447 | 0.7883 | 0.8098 | 0.7989 | 0.9789 |
0.0017 | 16.8998 | 14500 | 0.1483 | 0.7947 | 0.8025 | 0.7986 | 0.9791 |
0.0015 | 17.4825 | 15000 | 0.1508 | 0.8138 | 0.7918 | 0.8026 | 0.9795 |
0.0016 | 18.0653 | 15500 | 0.1463 | 0.7935 | 0.8093 | 0.8013 | 0.9796 |
0.0013 | 18.6480 | 16000 | 0.1506 | 0.7855 | 0.8085 | 0.7969 | 0.9791 |
0.0014 | 19.2308 | 16500 | 0.1521 | 0.7853 | 0.8143 | 0.7995 | 0.9790 |
0.0013 | 19.8135 | 17000 | 0.1580 | 0.7877 | 0.8061 | 0.7968 | 0.9787 |
0.0012 | 20.3963 | 17500 | 0.1557 | 0.7860 | 0.8182 | 0.8018 | 0.9795 |
0.0015 | 20.9790 | 18000 | 0.1576 | 0.7807 | 0.8192 | 0.7995 | 0.9786 |
0.001 | 21.5618 | 18500 | 0.1521 | 0.7972 | 0.8022 | 0.7997 | 0.9793 |
0.001 | 22.1445 | 19000 | 0.1542 | 0.7886 | 0.8038 | 0.7961 | 0.9791 |
0.0008 | 22.7273 | 19500 | 0.1579 | 0.7903 | 0.8116 | 0.8008 | 0.9795 |
0.0009 | 23.3100 | 20000 | 0.1592 | 0.7930 | 0.8075 | 0.8002 | 0.9795 |
0.0008 | 23.8928 | 20500 | 0.1596 | 0.7913 | 0.8081 | 0.7996 | 0.9794 |
0.0008 | 24.4755 | 21000 | 0.1567 | 0.7995 | 0.8085 | 0.8040 | 0.9798 |
0.0006 | 25.0583 | 21500 | 0.1603 | 0.7992 | 0.8005 | 0.7998 | 0.9794 |
0.0005 | 25.6410 | 22000 | 0.1610 | 0.7883 | 0.8171 | 0.8024 | 0.9795 |
0.0006 | 26.2238 | 22500 | 0.1620 | 0.7849 | 0.8136 | 0.7990 | 0.9791 |
0.0005 | 26.8065 | 23000 | 0.1623 | 0.7908 | 0.8108 | 0.8007 | 0.9791 |
0.0003 | 27.3893 | 23500 | 0.1643 | 0.7990 | 0.8022 | 0.8006 | 0.9795 |
0.0003 | 27.9720 | 24000 | 0.1628 | 0.7993 | 0.8080 | 0.8036 | 0.9796 |
0.0005 | 28.5548 | 24500 | 0.1640 | 0.7974 | 0.8093 | 0.8033 | 0.9795 |
0.0004 | 29.1375 | 25000 | 0.1640 | 0.7981 | 0.8085 | 0.8033 | 0.9795 |
0.0004 | 29.7203 | 25500 | 0.1641 | 0.7981 | 0.8067 | 0.8024 | 0.9795 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.1.1+cu121
- Datasets 2.14.5
- Tokenizers 0.19.1