Hugging Face's logo

haryoaw
/
scenario-non-kd-scr-ner-full-mdeberta_data-univner_full66

Token Classification
Transformers
Safetensors
deberta-v2
Generated from Trainer
Inference Endpoints
Model card Files Community
scenario-non-kd-scr-ner-full-mdeberta_data-univner_full66 / README.md
haryoaw's picture
haryoaw
Upload tokenizer
4702f33 verified
preview code
|
raw
history blame contribute delete
11.3 kB
metadata
base_model: microsoft/mdeberta-v3-base
library_name: transformers
license: mit
metrics:
  - precision
  - recall
  - f1
  - accuracy
tags:
  - generated_from_trainer
model-index:
  - name: scenario-non-kd-scr-ner-full-mdeberta_data-univner_full66
    results: []

scenario-non-kd-scr-ner-full-mdeberta_data-univner_full66

This model is a fine-tuned version of microsoft/mdeberta-v3-base on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3771
  • Precision: 0.6326
  • Recall: 0.6068
  • F1: 0.6194
  • Accuracy: 0.9635

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
0.3111 0.2910 500 0.2388 0.3312 0.2252 0.2681 0.9340
0.1965 0.5821 1000 0.2031 0.4073 0.2821 0.3333 0.9425
0.1493 0.8731 1500 0.1645 0.4478 0.4533 0.4505 0.9510
0.1123 1.1641 2000 0.1545 0.5054 0.5240 0.5146 0.9559
0.0879 1.4552 2500 0.1620 0.5659 0.4996 0.5307 0.9578
0.0809 1.7462 3000 0.1505 0.5293 0.5598 0.5441 0.9585
0.0722 2.0373 3500 0.1621 0.5762 0.5591 0.5675 0.9603
0.0457 2.3283 4000 0.1587 0.5676 0.5806 0.5740 0.9602
0.0478 2.6193 4500 0.1604 0.5524 0.5852 0.5683 0.9599
0.047 2.9104 5000 0.1550 0.5787 0.5865 0.5826 0.9612
0.0296 3.2014 5500 0.1791 0.5980 0.5989 0.5985 0.9622
0.0271 3.4924 6000 0.1783 0.6204 0.5865 0.6030 0.9623
0.0272 3.7835 6500 0.1794 0.5971 0.6074 0.6022 0.9620
0.0259 4.0745 7000 0.1968 0.6020 0.6136 0.6077 0.9627
0.0155 4.3655 7500 0.2028 0.5972 0.6053 0.6012 0.9626
0.0175 4.6566 8000 0.2095 0.6046 0.5754 0.5896 0.9615
0.0173 4.9476 8500 0.2121 0.5892 0.5931 0.5912 0.9619
0.0103 5.2386 9000 0.2274 0.6150 0.6024 0.6086 0.9627
0.0102 5.5297 9500 0.2231 0.6210 0.5905 0.6054 0.9623
0.0115 5.8207 10000 0.2175 0.6135 0.5966 0.6049 0.9624
0.0096 6.1118 10500 0.2394 0.5723 0.6358 0.6024 0.9613
0.0068 6.4028 11000 0.2474 0.6202 0.5957 0.6077 0.9629
0.007 6.6938 11500 0.2500 0.6095 0.6104 0.6100 0.9629
0.0085 6.9849 12000 0.2514 0.5995 0.5992 0.5994 0.9624
0.0054 7.2759 12500 0.2613 0.6161 0.5956 0.6057 0.9627
0.0052 7.5669 13000 0.2684 0.6083 0.6077 0.6080 0.9626
0.0056 7.8580 13500 0.2655 0.5795 0.6211 0.5996 0.9612
0.0048 8.1490 14000 0.2718 0.5925 0.6057 0.5990 0.9612
0.004 8.4400 14500 0.2794 0.6129 0.6094 0.6112 0.9624
0.0041 8.7311 15000 0.2811 0.6038 0.5937 0.5987 0.9618
0.0045 9.0221 15500 0.2814 0.6154 0.5878 0.6013 0.9622
0.0033 9.3132 16000 0.2879 0.5954 0.6203 0.6076 0.9621
0.0034 9.6042 16500 0.2963 0.6251 0.5956 0.6100 0.9631
0.0032 9.8952 17000 0.2935 0.5800 0.6321 0.6049 0.9615
0.0031 10.1863 17500 0.2909 0.6003 0.6194 0.6097 0.9625
0.0025 10.4773 18000 0.2991 0.5960 0.6096 0.6027 0.9619
0.0026 10.7683 18500 0.2983 0.6080 0.6086 0.6083 0.9623
0.0027 11.0594 19000 0.2975 0.6146 0.6054 0.6100 0.9624
0.0016 11.3504 19500 0.3092 0.6172 0.5900 0.6033 0.9626
0.0023 11.6414 20000 0.3168 0.6292 0.5918 0.6100 0.9630
0.0025 11.9325 20500 0.3036 0.6216 0.5972 0.6091 0.9627
0.0015 12.2235 21000 0.3222 0.6164 0.5918 0.6039 0.9621
0.0017 12.5146 21500 0.3158 0.6127 0.6089 0.6108 0.9626
0.0018 12.8056 22000 0.3223 0.6023 0.6008 0.6015 0.9623
0.0019 13.0966 22500 0.3197 0.6047 0.5910 0.5977 0.9618
0.0013 13.3877 23000 0.3190 0.6128 0.5985 0.6055 0.9620
0.0013 13.6787 23500 0.3279 0.6144 0.5904 0.6022 0.9622
0.0014 13.9697 24000 0.3278 0.6181 0.6089 0.6135 0.9624
0.0011 14.2608 24500 0.3384 0.6119 0.5927 0.6022 0.9623
0.0014 14.5518 25000 0.3270 0.6225 0.5993 0.6107 0.9621
0.0015 14.8428 25500 0.3320 0.5971 0.5969 0.5970 0.9616
0.001 15.1339 26000 0.3442 0.6174 0.5936 0.6053 0.9623
0.0008 15.4249 26500 0.3344 0.6091 0.6154 0.6122 0.9624
0.0009 15.7159 27000 0.3347 0.6242 0.5982 0.6109 0.9625
0.0011 16.0070 27500 0.3407 0.6225 0.6064 0.6143 0.9625
0.0008 16.2980 28000 0.3376 0.6217 0.6081 0.6148 0.9626
0.0008 16.5891 28500 0.3476 0.6030 0.6130 0.6080 0.9627
0.0009 16.8801 29000 0.3390 0.6224 0.5988 0.6103 0.9626
0.0009 17.1711 29500 0.3427 0.6094 0.6195 0.6144 0.9624
0.0006 17.4622 30000 0.3451 0.6200 0.6126 0.6163 0.9629
0.0004 17.7532 30500 0.3485 0.6190 0.6078 0.6134 0.9630
0.0008 18.0442 31000 0.3532 0.6237 0.5973 0.6102 0.9628
0.0007 18.3353 31500 0.3454 0.6143 0.6019 0.6080 0.9628
0.0006 18.6263 32000 0.3426 0.6253 0.6093 0.6172 0.9629
0.0006 18.9173 32500 0.3503 0.6205 0.6018 0.6110 0.9628
0.0004 19.2084 33000 0.3580 0.6344 0.6034 0.6185 0.9633
0.0004 19.4994 33500 0.3527 0.6072 0.6203 0.6137 0.9626
0.0006 19.7905 34000 0.3473 0.6173 0.6115 0.6144 0.9628
0.0005 20.0815 34500 0.3550 0.6208 0.6106 0.6157 0.9630
0.0003 20.3725 35000 0.3623 0.6153 0.6074 0.6113 0.9626
0.0004 20.6636 35500 0.3639 0.6264 0.5989 0.6123 0.9628
0.0005 20.9546 36000 0.3505 0.6167 0.6179 0.6173 0.9631
0.0004 21.2456 36500 0.3570 0.6237 0.6093 0.6164 0.9631
0.0003 21.5367 37000 0.3608 0.6302 0.6089 0.6194 0.9634
0.0005 21.8277 37500 0.3597 0.6158 0.6027 0.6092 0.9626
0.0002 22.1187 38000 0.3595 0.6252 0.6070 0.6160 0.9632
0.0003 22.4098 38500 0.3615 0.6186 0.6135 0.6160 0.9631
0.0002 22.7008 39000 0.3630 0.6311 0.5983 0.6143 0.9633
0.0003 22.9919 39500 0.3694 0.6344 0.5825 0.6073 0.9629
0.0001 23.2829 40000 0.3673 0.6284 0.6071 0.6176 0.9634
0.0002 23.5739 40500 0.3693 0.6187 0.6063 0.6124 0.9630
0.0003 23.8650 41000 0.3704 0.6153 0.6087 0.6120 0.9630
0.0001 24.1560 41500 0.3663 0.6219 0.6070 0.6143 0.9633
0.0001 24.4470 42000 0.3667 0.6228 0.6161 0.6194 0.9637
0.0002 24.7381 42500 0.3736 0.6456 0.5926 0.6179 0.9633
0.0002 25.0291 43000 0.3742 0.6280 0.5953 0.6112 0.9633
0.0001 25.3201 43500 0.3714 0.6217 0.6016 0.6115 0.9629
0.0002 25.6112 44000 0.3720 0.6348 0.5933 0.6133 0.9632
0.0001 25.9022 44500 0.3726 0.6136 0.6152 0.6144 0.9631
0.0001 26.1932 45000 0.3694 0.6366 0.5963 0.6158 0.9636
0.0001 26.4843 45500 0.3678 0.6113 0.6227 0.6170 0.9632
0.0001 26.7753 46000 0.3702 0.6348 0.6016 0.6178 0.9636
0.0001 27.0664 46500 0.3747 0.6323 0.5998 0.6156 0.9634
0.0001 27.3574 47000 0.3738 0.6352 0.6006 0.6174 0.9635
0.0001 27.6484 47500 0.3701 0.6215 0.6135 0.6174 0.9633
0.0001 27.9395 48000 0.3718 0.6252 0.6122 0.6186 0.9633
0.0001 28.2305 48500 0.3743 0.6308 0.6066 0.6184 0.9634
0.0 28.5215 49000 0.3785 0.6333 0.5957 0.6139 0.9634
0.0001 28.8126 49500 0.3764 0.6258 0.6087 0.6171 0.9633
0.0001 29.1036 50000 0.3761 0.6266 0.6103 0.6183 0.9634
0.0001 29.3946 50500 0.3770 0.6333 0.6051 0.6189 0.9634
0.0001 29.6857 51000 0.3780 0.6346 0.6037 0.6188 0.9635
0.0 29.9767 51500 0.3771 0.6326 0.6068 0.6194 0.9635

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.1.1+cu121
  • Datasets 2.14.5
  • Tokenizers 0.19.1