--- library_name: transformers license: mit base_model: microsoft/mdeberta-v3-base tags: - generated_from_trainer metrics: - precision - recall - f1 - accuracy model-index: - name: scenario-non-kd-pre-ner-full-mdeberta_data-univner_full55 results: [] --- # scenario-non-kd-pre-ner-full-mdeberta_data-univner_full55 This model is a fine-tuned version of [microsoft/mdeberta-v3-base](https://huggingface.co/microsoft/mdeberta-v3-base) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.1841 - Precision: 0.8015 - Recall: 0.8248 - F1: 0.8130 - Accuracy: 0.9804 ## 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: 55 - 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 | |:-------------:|:-------:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:| | 0.2548 | 0.2910 | 500 | 0.1794 | 0.3392 | 0.4613 | 0.3909 | 0.9392 | | 0.1474 | 0.5821 | 1000 | 0.1201 | 0.5290 | 0.6425 | 0.5802 | 0.9597 | | 0.0973 | 0.8731 | 1500 | 0.1020 | 0.5970 | 0.7570 | 0.6676 | 0.9651 | | 0.0751 | 1.1641 | 2000 | 0.0905 | 0.6815 | 0.7547 | 0.7162 | 0.9719 | | 0.062 | 1.4552 | 2500 | 0.0891 | 0.6689 | 0.7826 | 0.7213 | 0.9706 | | 0.0566 | 1.7462 | 3000 | 0.0927 | 0.6774 | 0.8134 | 0.7392 | 0.9726 | | 0.0506 | 2.0373 | 3500 | 0.0850 | 0.7449 | 0.7879 | 0.7658 | 0.9766 | | 0.0407 | 2.3283 | 4000 | 0.0832 | 0.7512 | 0.7969 | 0.7734 | 0.9775 | | 0.0374 | 2.6193 | 4500 | 0.0828 | 0.7646 | 0.7974 | 0.7806 | 0.9781 | | 0.036 | 2.9104 | 5000 | 0.0852 | 0.7362 | 0.8211 | 0.7763 | 0.9766 | | 0.0277 | 3.2014 | 5500 | 0.0903 | 0.7474 | 0.8104 | 0.7777 | 0.9773 | | 0.0254 | 3.4924 | 6000 | 0.0928 | 0.7448 | 0.8149 | 0.7783 | 0.9769 | | 0.0254 | 3.7835 | 6500 | 0.0881 | 0.7661 | 0.8038 | 0.7845 | 0.9789 | | 0.0229 | 4.0745 | 7000 | 0.0976 | 0.7564 | 0.8204 | 0.7871 | 0.9784 | | 0.017 | 4.3655 | 7500 | 0.0949 | 0.7647 | 0.8116 | 0.7874 | 0.9784 | | 0.018 | 4.6566 | 8000 | 0.0992 | 0.7693 | 0.8035 | 0.7860 | 0.9779 | | 0.0183 | 4.9476 | 8500 | 0.1043 | 0.7392 | 0.8373 | 0.7852 | 0.9768 | | 0.014 | 5.2386 | 9000 | 0.1032 | 0.7642 | 0.8221 | 0.7921 | 0.9783 | | 0.0127 | 5.5297 | 9500 | 0.1045 | 0.7600 | 0.8256 | 0.7914 | 0.9780 | | 0.0133 | 5.8207 | 10000 | 0.1071 | 0.7724 | 0.8015 | 0.7867 | 0.9780 | | 0.013 | 6.1118 | 10500 | 0.1089 | 0.7804 | 0.8129 | 0.7963 | 0.9792 | | 0.0101 | 6.4028 | 11000 | 0.1110 | 0.7918 | 0.8071 | 0.7994 | 0.9790 | | 0.0103 | 6.6938 | 11500 | 0.1102 | 0.7761 | 0.8178 | 0.7964 | 0.9783 | | 0.0109 | 6.9849 | 12000 | 0.1082 | 0.7658 | 0.8273 | 0.7953 | 0.9781 | | 0.0077 | 7.2759 | 12500 | 0.1196 | 0.7627 | 0.8188 | 0.7897 | 0.9784 | | 0.008 | 7.5669 | 13000 | 0.1187 | 0.7664 | 0.8181 | 0.7914 | 0.9783 | | 0.0087 | 7.8580 | 13500 | 0.1136 | 0.7819 | 0.8179 | 0.7995 | 0.9791 | | 0.0082 | 8.1490 | 14000 | 0.1227 | 0.7753 | 0.8178 | 0.7960 | 0.9792 | | 0.0061 | 8.4400 | 14500 | 0.1244 | 0.7782 | 0.8207 | 0.7989 | 0.9789 | | 0.0067 | 8.7311 | 15000 | 0.1293 | 0.7606 | 0.8270 | 0.7924 | 0.9780 | | 0.0058 | 9.0221 | 15500 | 0.1379 | 0.7671 | 0.8145 | 0.7901 | 0.9775 | | 0.0056 | 9.3132 | 16000 | 0.1311 | 0.7873 | 0.8113 | 0.7991 | 0.9794 | | 0.0051 | 9.6042 | 16500 | 0.1273 | 0.7852 | 0.8084 | 0.7966 | 0.9794 | | 0.0048 | 9.8952 | 17000 | 0.1395 | 0.7557 | 0.8309 | 0.7915 | 0.9773 | | 0.0043 | 10.1863 | 17500 | 0.1349 | 0.7967 | 0.8165 | 0.8065 | 0.9797 | | 0.0042 | 10.4773 | 18000 | 0.1322 | 0.7949 | 0.8110 | 0.8029 | 0.9795 | | 0.0043 | 10.7683 | 18500 | 0.1372 | 0.7900 | 0.8087 | 0.7992 | 0.9792 | | 0.0046 | 11.0594 | 19000 | 0.1335 | 0.8052 | 0.8012 | 0.8032 | 0.9797 | | 0.0032 | 11.3504 | 19500 | 0.1388 | 0.7785 | 0.8238 | 0.8005 | 0.9792 | | 0.0036 | 11.6414 | 20000 | 0.1454 | 0.7869 | 0.8221 | 0.8041 | 0.9795 | | 0.004 | 11.9325 | 20500 | 0.1351 | 0.7869 | 0.8097 | 0.7981 | 0.9789 | | 0.0029 | 12.2235 | 21000 | 0.1549 | 0.7811 | 0.8238 | 0.8019 | 0.9791 | | 0.0025 | 12.5146 | 21500 | 0.1498 | 0.7914 | 0.8160 | 0.8035 | 0.9794 | | 0.0031 | 12.8056 | 22000 | 0.1470 | 0.7946 | 0.8142 | 0.8042 | 0.9793 | | 0.0034 | 13.0966 | 22500 | 0.1481 | 0.7878 | 0.8137 | 0.8006 | 0.9793 | | 0.0022 | 13.3877 | 23000 | 0.1500 | 0.7884 | 0.8225 | 0.8051 | 0.9794 | | 0.0026 | 13.6787 | 23500 | 0.1525 | 0.7870 | 0.8277 | 0.8068 | 0.9796 | | 0.003 | 13.9697 | 24000 | 0.1482 | 0.7918 | 0.8205 | 0.8059 | 0.9798 | | 0.0021 | 14.2608 | 24500 | 0.1533 | 0.7786 | 0.8234 | 0.8004 | 0.9792 | | 0.0023 | 14.5518 | 25000 | 0.1453 | 0.7835 | 0.8238 | 0.8032 | 0.9797 | | 0.0024 | 14.8428 | 25500 | 0.1506 | 0.7873 | 0.8277 | 0.8070 | 0.9796 | | 0.0021 | 15.1339 | 26000 | 0.1561 | 0.7996 | 0.8169 | 0.8082 | 0.9798 | | 0.0018 | 15.4249 | 26500 | 0.1603 | 0.7859 | 0.8243 | 0.8046 | 0.9793 | | 0.0019 | 15.7159 | 27000 | 0.1665 | 0.7835 | 0.8189 | 0.8008 | 0.9794 | | 0.0019 | 16.0070 | 27500 | 0.1543 | 0.7921 | 0.8153 | 0.8036 | 0.9796 | | 0.0017 | 16.2980 | 28000 | 0.1694 | 0.7956 | 0.8130 | 0.8042 | 0.9796 | | 0.0018 | 16.5891 | 28500 | 0.1573 | 0.7930 | 0.8100 | 0.8014 | 0.9794 | | 0.0019 | 16.8801 | 29000 | 0.1544 | 0.7930 | 0.8220 | 0.8072 | 0.9795 | | 0.0016 | 17.1711 | 29500 | 0.1628 | 0.7900 | 0.8237 | 0.8065 | 0.9795 | | 0.0014 | 17.4622 | 30000 | 0.1609 | 0.7875 | 0.8208 | 0.8038 | 0.9793 | | 0.0014 | 17.7532 | 30500 | 0.1637 | 0.7844 | 0.8241 | 0.8038 | 0.9793 | | 0.0016 | 18.0442 | 31000 | 0.1647 | 0.7964 | 0.8124 | 0.8043 | 0.9798 | | 0.0014 | 18.3353 | 31500 | 0.1690 | 0.7962 | 0.8149 | 0.8054 | 0.9800 | | 0.0012 | 18.6263 | 32000 | 0.1759 | 0.7837 | 0.8254 | 0.8040 | 0.9792 | | 0.001 | 18.9173 | 32500 | 0.1764 | 0.7950 | 0.8159 | 0.8053 | 0.9797 | | 0.0011 | 19.2084 | 33000 | 0.1684 | 0.8022 | 0.8106 | 0.8064 | 0.9802 | | 0.001 | 19.4994 | 33500 | 0.1723 | 0.7944 | 0.8143 | 0.8042 | 0.9797 | | 0.0014 | 19.7905 | 34000 | 0.1637 | 0.7886 | 0.8300 | 0.8088 | 0.9801 | | 0.0013 | 20.0815 | 34500 | 0.1633 | 0.7885 | 0.8201 | 0.8040 | 0.9799 | | 0.001 | 20.3725 | 35000 | 0.1682 | 0.7955 | 0.8251 | 0.8101 | 0.9799 | | 0.0009 | 20.6636 | 35500 | 0.1735 | 0.7956 | 0.8222 | 0.8087 | 0.9798 | | 0.0009 | 20.9546 | 36000 | 0.1765 | 0.7898 | 0.8231 | 0.8061 | 0.9796 | | 0.0012 | 21.2456 | 36500 | 0.1754 | 0.7977 | 0.8243 | 0.8108 | 0.9800 | | 0.0006 | 21.5367 | 37000 | 0.1710 | 0.7978 | 0.8176 | 0.8076 | 0.9799 | | 0.0008 | 21.8277 | 37500 | 0.1650 | 0.8035 | 0.8256 | 0.8144 | 0.9800 | | 0.0007 | 22.1187 | 38000 | 0.1702 | 0.7893 | 0.8341 | 0.8111 | 0.9795 | | 0.0007 | 22.4098 | 38500 | 0.1724 | 0.8086 | 0.8067 | 0.8077 | 0.9801 | | 0.0008 | 22.7008 | 39000 | 0.1714 | 0.7951 | 0.8224 | 0.8085 | 0.9800 | | 0.0007 | 22.9919 | 39500 | 0.1757 | 0.8010 | 0.8194 | 0.8101 | 0.9803 | | 0.0005 | 23.2829 | 40000 | 0.1776 | 0.8002 | 0.8261 | 0.8129 | 0.9802 | | 0.0007 | 23.5739 | 40500 | 0.1771 | 0.7978 | 0.8273 | 0.8123 | 0.9802 | | 0.0007 | 23.8650 | 41000 | 0.1800 | 0.7984 | 0.8233 | 0.8106 | 0.9801 | | 0.0007 | 24.1560 | 41500 | 0.1761 | 0.7970 | 0.8261 | 0.8113 | 0.9799 | | 0.0005 | 24.4470 | 42000 | 0.1729 | 0.8041 | 0.8220 | 0.8129 | 0.9801 | | 0.0007 | 24.7381 | 42500 | 0.1753 | 0.8066 | 0.8214 | 0.8139 | 0.9802 | | 0.0004 | 25.0291 | 43000 | 0.1812 | 0.7875 | 0.8308 | 0.8085 | 0.9796 | | 0.0006 | 25.3201 | 43500 | 0.1773 | 0.7991 | 0.8240 | 0.8113 | 0.9801 | | 0.0005 | 25.6112 | 44000 | 0.1771 | 0.7996 | 0.8168 | 0.8081 | 0.9801 | | 0.0005 | 25.9022 | 44500 | 0.1780 | 0.7986 | 0.8266 | 0.8123 | 0.9801 | | 0.0004 | 26.1932 | 45000 | 0.1788 | 0.7999 | 0.8227 | 0.8112 | 0.9802 | | 0.0005 | 26.4843 | 45500 | 0.1792 | 0.7981 | 0.8277 | 0.8127 | 0.9802 | | 0.0004 | 26.7753 | 46000 | 0.1807 | 0.7959 | 0.8250 | 0.8102 | 0.9801 | | 0.0004 | 27.0664 | 46500 | 0.1807 | 0.8079 | 0.8217 | 0.8147 | 0.9804 | | 0.0005 | 27.3574 | 47000 | 0.1818 | 0.8013 | 0.8254 | 0.8132 | 0.9803 | | 0.0005 | 27.6484 | 47500 | 0.1814 | 0.7985 | 0.8220 | 0.8100 | 0.9802 | | 0.0003 | 27.9395 | 48000 | 0.1831 | 0.8010 | 0.8261 | 0.8134 | 0.9803 | | 0.0003 | 28.2305 | 48500 | 0.1836 | 0.8051 | 0.8222 | 0.8136 | 0.9803 | | 0.0002 | 28.5215 | 49000 | 0.1857 | 0.8028 | 0.8237 | 0.8131 | 0.9803 | | 0.0007 | 28.8126 | 49500 | 0.1839 | 0.7976 | 0.8272 | 0.8121 | 0.9803 | | 0.0003 | 29.1036 | 50000 | 0.1839 | 0.8037 | 0.8243 | 0.8139 | 0.9803 | | 0.0002 | 29.3946 | 50500 | 0.1842 | 0.8026 | 0.8238 | 0.8131 | 0.9804 | | 0.0002 | 29.6857 | 51000 | 0.1843 | 0.8004 | 0.8247 | 0.8124 | 0.9803 | | 0.0003 | 29.9767 | 51500 | 0.1841 | 0.8015 | 0.8248 | 0.8130 | 0.9804 | ### Framework versions - Transformers 4.44.2 - Pytorch 2.1.1+cu121 - Datasets 2.14.5 - Tokenizers 0.19.1