--- 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-scr-ner-half-mdeberta_data-univner_full44 results: [] --- # scenario-non-kd-scr-ner-half-mdeberta_data-univner_full44 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: 0.3534 - Precision: 0.6234 - Recall: 0.5829 - F1: 0.6024 - Accuracy: 0.9614 ## 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 | |:-------------:|:-------:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:| | 0.3562 | 0.2910 | 500 | 0.2852 | 0.2966 | 0.1133 | 0.1639 | 0.9283 | | 0.2364 | 0.5821 | 1000 | 0.2099 | 0.3403 | 0.2622 | 0.2961 | 0.9384 | | 0.1729 | 0.8731 | 1500 | 0.1788 | 0.4217 | 0.3849 | 0.4025 | 0.9465 | | 0.1341 | 1.1641 | 2000 | 0.1651 | 0.4680 | 0.4595 | 0.4637 | 0.9513 | | 0.1123 | 1.4552 | 2500 | 0.1575 | 0.5014 | 0.4913 | 0.4963 | 0.9539 | | 0.104 | 1.7462 | 3000 | 0.1514 | 0.5253 | 0.5071 | 0.5161 | 0.9559 | | 0.0921 | 2.0373 | 3500 | 0.1484 | 0.5065 | 0.5485 | 0.5267 | 0.9561 | | 0.0674 | 2.3283 | 4000 | 0.1515 | 0.5461 | 0.5643 | 0.5550 | 0.9584 | | 0.0661 | 2.6193 | 4500 | 0.1523 | 0.5477 | 0.5588 | 0.5532 | 0.9589 | | 0.0649 | 2.9104 | 5000 | 0.1511 | 0.5704 | 0.5721 | 0.5712 | 0.9600 | | 0.0493 | 3.2014 | 5500 | 0.1632 | 0.5742 | 0.5666 | 0.5704 | 0.9601 | | 0.0441 | 3.4924 | 6000 | 0.1626 | 0.5910 | 0.5748 | 0.5828 | 0.9605 | | 0.0446 | 3.7835 | 6500 | 0.1678 | 0.6082 | 0.5561 | 0.5809 | 0.9611 | | 0.0396 | 4.0745 | 7000 | 0.1740 | 0.5672 | 0.5738 | 0.5705 | 0.9601 | | 0.0312 | 4.3655 | 7500 | 0.1776 | 0.5913 | 0.5713 | 0.5812 | 0.9605 | | 0.0311 | 4.6566 | 8000 | 0.1723 | 0.5811 | 0.5944 | 0.5877 | 0.9601 | | 0.0306 | 4.9476 | 8500 | 0.1793 | 0.5924 | 0.5807 | 0.5865 | 0.9608 | | 0.0221 | 5.2386 | 9000 | 0.1898 | 0.5809 | 0.5931 | 0.5870 | 0.9605 | | 0.0214 | 5.5297 | 9500 | 0.1910 | 0.5876 | 0.5827 | 0.5852 | 0.9602 | | 0.0229 | 5.8207 | 10000 | 0.1947 | 0.5694 | 0.5934 | 0.5811 | 0.9596 | | 0.0203 | 6.1118 | 10500 | 0.2044 | 0.6051 | 0.5732 | 0.5887 | 0.9608 | | 0.015 | 6.4028 | 11000 | 0.2076 | 0.5915 | 0.5954 | 0.5935 | 0.9606 | | 0.0157 | 6.6938 | 11500 | 0.2152 | 0.5937 | 0.5853 | 0.5895 | 0.9603 | | 0.0166 | 6.9849 | 12000 | 0.2135 | 0.6049 | 0.5826 | 0.5935 | 0.9606 | | 0.0115 | 7.2759 | 12500 | 0.2225 | 0.5847 | 0.5887 | 0.5867 | 0.9599 | | 0.0119 | 7.5669 | 13000 | 0.2173 | 0.6006 | 0.5996 | 0.6001 | 0.9606 | | 0.0124 | 7.8580 | 13500 | 0.2253 | 0.6116 | 0.5855 | 0.5983 | 0.9613 | | 0.0099 | 8.1490 | 14000 | 0.2324 | 0.6004 | 0.5876 | 0.5939 | 0.9613 | | 0.0081 | 8.4400 | 14500 | 0.2409 | 0.6121 | 0.5729 | 0.5918 | 0.9611 | | 0.0097 | 8.7311 | 15000 | 0.2405 | 0.5896 | 0.5809 | 0.5852 | 0.9600 | | 0.0088 | 9.0221 | 15500 | 0.2459 | 0.5980 | 0.5866 | 0.5923 | 0.9609 | | 0.0065 | 9.3132 | 16000 | 0.2465 | 0.6075 | 0.5866 | 0.5969 | 0.9612 | | 0.0069 | 9.6042 | 16500 | 0.2520 | 0.6127 | 0.5709 | 0.5911 | 0.9609 | | 0.0073 | 9.8952 | 17000 | 0.2520 | 0.6088 | 0.5851 | 0.5967 | 0.9613 | | 0.0062 | 10.1863 | 17500 | 0.2605 | 0.6225 | 0.5599 | 0.5895 | 0.9608 | | 0.0047 | 10.4773 | 18000 | 0.2612 | 0.6002 | 0.5806 | 0.5902 | 0.9607 | | 0.0054 | 10.7683 | 18500 | 0.2615 | 0.6137 | 0.5833 | 0.5981 | 0.9611 | | 0.0053 | 11.0594 | 19000 | 0.2658 | 0.6199 | 0.5838 | 0.6013 | 0.9614 | | 0.0045 | 11.3504 | 19500 | 0.2701 | 0.6195 | 0.5813 | 0.5998 | 0.9615 | | 0.0047 | 11.6414 | 20000 | 0.2753 | 0.6101 | 0.5744 | 0.5917 | 0.9610 | | 0.004 | 11.9325 | 20500 | 0.2671 | 0.6004 | 0.5898 | 0.5951 | 0.9605 | | 0.004 | 12.2235 | 21000 | 0.2692 | 0.6134 | 0.5849 | 0.5988 | 0.9615 | | 0.0035 | 12.5146 | 21500 | 0.2752 | 0.6215 | 0.5833 | 0.6018 | 0.9615 | | 0.0038 | 12.8056 | 22000 | 0.2782 | 0.6198 | 0.5709 | 0.5944 | 0.9616 | | 0.0032 | 13.0966 | 22500 | 0.2833 | 0.6040 | 0.5820 | 0.5928 | 0.9611 | | 0.0029 | 13.3877 | 23000 | 0.2855 | 0.6122 | 0.5698 | 0.5902 | 0.9605 | | 0.0029 | 13.6787 | 23500 | 0.2882 | 0.6066 | 0.5859 | 0.5961 | 0.9613 | | 0.0031 | 13.9697 | 24000 | 0.2927 | 0.6072 | 0.5788 | 0.5927 | 0.9608 | | 0.002 | 14.2608 | 24500 | 0.2950 | 0.6220 | 0.5739 | 0.5970 | 0.9613 | | 0.0024 | 14.5518 | 25000 | 0.2941 | 0.6104 | 0.5830 | 0.5964 | 0.9612 | | 0.0026 | 14.8428 | 25500 | 0.2932 | 0.6181 | 0.5865 | 0.6019 | 0.9617 | | 0.002 | 15.1339 | 26000 | 0.3020 | 0.6059 | 0.5827 | 0.5941 | 0.9610 | | 0.0019 | 15.4249 | 26500 | 0.3010 | 0.6254 | 0.5807 | 0.6022 | 0.9616 | | 0.0024 | 15.7159 | 27000 | 0.3093 | 0.6379 | 0.5563 | 0.5943 | 0.9613 | | 0.002 | 16.0070 | 27500 | 0.3038 | 0.5999 | 0.5953 | 0.5976 | 0.9611 | | 0.0015 | 16.2980 | 28000 | 0.3101 | 0.6056 | 0.6014 | 0.6035 | 0.9616 | | 0.0019 | 16.5891 | 28500 | 0.3110 | 0.6152 | 0.5742 | 0.5940 | 0.9610 | | 0.0018 | 16.8801 | 29000 | 0.3143 | 0.6179 | 0.5842 | 0.6006 | 0.9612 | | 0.0016 | 17.1711 | 29500 | 0.3179 | 0.6280 | 0.5799 | 0.6030 | 0.9614 | | 0.0013 | 17.4622 | 30000 | 0.3202 | 0.6165 | 0.5778 | 0.5966 | 0.9615 | | 0.0016 | 17.7532 | 30500 | 0.3185 | 0.6162 | 0.5879 | 0.6017 | 0.9614 | | 0.0012 | 18.0442 | 31000 | 0.3236 | 0.6151 | 0.5784 | 0.5962 | 0.9614 | | 0.0009 | 18.3353 | 31500 | 0.3210 | 0.6160 | 0.5920 | 0.6037 | 0.9616 | | 0.0013 | 18.6263 | 32000 | 0.3265 | 0.6257 | 0.5750 | 0.5992 | 0.9613 | | 0.0013 | 18.9173 | 32500 | 0.3219 | 0.6199 | 0.5778 | 0.5981 | 0.9612 | | 0.0013 | 19.2084 | 33000 | 0.3215 | 0.6142 | 0.5839 | 0.5987 | 0.9614 | | 0.0011 | 19.4994 | 33500 | 0.3180 | 0.6189 | 0.5891 | 0.6036 | 0.9616 | | 0.0011 | 19.7905 | 34000 | 0.3217 | 0.6192 | 0.5879 | 0.6032 | 0.9615 | | 0.0009 | 20.0815 | 34500 | 0.3240 | 0.6018 | 0.5979 | 0.5998 | 0.9612 | | 0.0012 | 20.3725 | 35000 | 0.3250 | 0.6120 | 0.5904 | 0.6010 | 0.9611 | | 0.001 | 20.6636 | 35500 | 0.3277 | 0.6196 | 0.5851 | 0.6018 | 0.9615 | | 0.0009 | 20.9546 | 36000 | 0.3354 | 0.6251 | 0.5729 | 0.5979 | 0.9614 | | 0.0008 | 21.2456 | 36500 | 0.3315 | 0.6177 | 0.5783 | 0.5973 | 0.9613 | | 0.0008 | 21.5367 | 37000 | 0.3258 | 0.6185 | 0.5875 | 0.6026 | 0.9617 | | 0.0008 | 21.8277 | 37500 | 0.3327 | 0.6236 | 0.5842 | 0.6032 | 0.9619 | | 0.0008 | 22.1187 | 38000 | 0.3309 | 0.6071 | 0.6015 | 0.6043 | 0.9614 | | 0.0006 | 22.4098 | 38500 | 0.3401 | 0.6302 | 0.5638 | 0.5952 | 0.9613 | | 0.0006 | 22.7008 | 39000 | 0.3372 | 0.6285 | 0.5787 | 0.6026 | 0.9617 | | 0.0008 | 22.9919 | 39500 | 0.3391 | 0.6189 | 0.5855 | 0.6017 | 0.9615 | | 0.0007 | 23.2829 | 40000 | 0.3356 | 0.6190 | 0.5874 | 0.6028 | 0.9619 | | 0.0005 | 23.5739 | 40500 | 0.3330 | 0.6222 | 0.5868 | 0.6040 | 0.9620 | | 0.0009 | 23.8650 | 41000 | 0.3381 | 0.6156 | 0.5846 | 0.5997 | 0.9610 | | 0.0004 | 24.1560 | 41500 | 0.3460 | 0.6298 | 0.5732 | 0.6002 | 0.9614 | | 0.0005 | 24.4470 | 42000 | 0.3442 | 0.6215 | 0.5881 | 0.6043 | 0.9615 | | 0.0006 | 24.7381 | 42500 | 0.3467 | 0.6240 | 0.5848 | 0.6038 | 0.9617 | | 0.0005 | 25.0291 | 43000 | 0.3492 | 0.6307 | 0.5734 | 0.6007 | 0.9615 | | 0.0006 | 25.3201 | 43500 | 0.3411 | 0.6287 | 0.5823 | 0.6046 | 0.9619 | | 0.0003 | 25.6112 | 44000 | 0.3486 | 0.6342 | 0.5705 | 0.6006 | 0.9616 | | 0.0004 | 25.9022 | 44500 | 0.3437 | 0.6257 | 0.5817 | 0.6029 | 0.9614 | | 0.0005 | 26.1932 | 45000 | 0.3434 | 0.6152 | 0.5891 | 0.6019 | 0.9614 | | 0.0003 | 26.4843 | 45500 | 0.3486 | 0.6239 | 0.5755 | 0.5987 | 0.9612 | | 0.0005 | 26.7753 | 46000 | 0.3492 | 0.6083 | 0.5869 | 0.5974 | 0.9613 | | 0.0004 | 27.0664 | 46500 | 0.3531 | 0.6198 | 0.5767 | 0.5975 | 0.9612 | | 0.0003 | 27.3574 | 47000 | 0.3489 | 0.6178 | 0.5874 | 0.6022 | 0.9613 | | 0.0004 | 27.6484 | 47500 | 0.3489 | 0.6184 | 0.5839 | 0.6007 | 0.9612 | | 0.0003 | 27.9395 | 48000 | 0.3517 | 0.6191 | 0.5813 | 0.5996 | 0.9613 | | 0.0003 | 28.2305 | 48500 | 0.3523 | 0.6227 | 0.5806 | 0.6009 | 0.9614 | | 0.0002 | 28.5215 | 49000 | 0.3530 | 0.6225 | 0.5852 | 0.6033 | 0.9615 | | 0.0003 | 28.8126 | 49500 | 0.3528 | 0.6234 | 0.5820 | 0.6020 | 0.9614 | | 0.0003 | 29.1036 | 50000 | 0.3531 | 0.6205 | 0.5830 | 0.6012 | 0.9613 | | 0.0004 | 29.3946 | 50500 | 0.3521 | 0.6202 | 0.5875 | 0.6034 | 0.9614 | | 0.0003 | 29.6857 | 51000 | 0.3532 | 0.6219 | 0.5836 | 0.6022 | 0.9613 | | 0.0003 | 29.9767 | 51500 | 0.3534 | 0.6234 | 0.5829 | 0.6024 | 0.9614 | ### Framework versions - Transformers 4.44.2 - Pytorch 2.1.1+cu121 - Datasets 2.14.5 - Tokenizers 0.19.1