Edit model card

bert_multi-uncased-finetuned-pos-tr

This model is a fine-tuned version of bert-base-multilingual-uncased on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3933
  • Precision: 0.9225
  • Recall: 0.9275
  • F1: 0.9250
  • Accuracy: 0.9564

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
  • num_epochs: 50

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
0.4051 1.0 836 0.1752 0.9110 0.9098 0.9104 0.9493
0.1506 2.0 1672 0.1492 0.9112 0.9207 0.9159 0.9529
0.1182 3.0 2508 0.1632 0.9183 0.9229 0.9206 0.9552
0.0973 4.0 3344 0.1544 0.9197 0.9216 0.9206 0.9550
0.0796 5.0 4180 0.1676 0.9227 0.9221 0.9224 0.9569
0.0673 6.0 5016 0.1829 0.9196 0.9206 0.9201 0.9548
0.052 7.0 5852 0.1850 0.9220 0.9237 0.9228 0.9562
0.0448 8.0 6688 0.1980 0.9187 0.9305 0.9246 0.9571
0.0394 9.0 7524 0.2298 0.9147 0.9204 0.9176 0.9531
0.0298 10.0 8360 0.2288 0.9217 0.9277 0.9247 0.9571
0.028 11.0 9196 0.2434 0.9173 0.9231 0.9202 0.9545
0.0233 12.0 10032 0.2609 0.9174 0.9275 0.9225 0.9558
0.018 13.0 10868 0.2701 0.9211 0.9249 0.9230 0.9559
0.0168 14.0 11704 0.2694 0.9206 0.9261 0.9234 0.9564
0.0157 15.0 12540 0.2774 0.9213 0.929 0.9251 0.9568
0.0148 16.0 13376 0.2923 0.9229 0.9255 0.9242 0.9565
0.013 17.0 14212 0.2853 0.9229 0.9263 0.9246 0.9574
0.0104 18.0 15048 0.2899 0.9224 0.9265 0.9245 0.9569
0.0096 19.0 15884 0.2769 0.9239 0.9253 0.9246 0.9571
0.009 20.0 16720 0.3074 0.9204 0.9267 0.9235 0.9570
0.0098 21.0 17556 0.2985 0.9221 0.9287 0.9254 0.9573
0.0071 22.0 18392 0.3293 0.9216 0.9277 0.9246 0.9561
0.0069 23.0 19228 0.3135 0.9242 0.9285 0.9263 0.9572
0.007 24.0 20064 0.3098 0.9226 0.9291 0.9258 0.9576
0.0072 25.0 20900 0.3352 0.9241 0.9288 0.9265 0.9580
0.0048 26.0 21736 0.3384 0.9228 0.9269 0.9249 0.9567
0.006 27.0 22572 0.3316 0.9232 0.9276 0.9254 0.9575
0.0057 28.0 23408 0.3381 0.9238 0.9272 0.9255 0.9578
0.0051 29.0 24244 0.3494 0.9211 0.9282 0.9246 0.9568
0.0046 30.0 25080 0.3379 0.9237 0.9254 0.9246 0.9570
0.0044 31.0 25916 0.3512 0.9232 0.9251 0.9242 0.9565
0.0037 32.0 26752 0.3625 0.9227 0.9252 0.9240 0.9563
0.0029 33.0 27588 0.3476 0.9220 0.9264 0.9242 0.9574
0.0039 34.0 28424 0.3635 0.9238 0.9275 0.9257 0.9575
0.0034 35.0 29260 0.3685 0.9205 0.9247 0.9226 0.9554
0.0033 36.0 30096 0.3693 0.9219 0.9245 0.9232 0.9555
0.003 37.0 30932 0.3698 0.9239 0.9257 0.9248 0.9573
0.0024 38.0 31768 0.3772 0.9242 0.926 0.9251 0.9570
0.0029 39.0 32604 0.3798 0.9246 0.9281 0.9263 0.9563
0.0024 40.0 33440 0.3804 0.9215 0.9264 0.9239 0.9562
0.0017 41.0 34276 0.3804 0.9238 0.9274 0.9256 0.9570
0.0025 42.0 35112 0.3808 0.9252 0.9273 0.9263 0.9570
0.0025 43.0 35948 0.3794 0.9237 0.9282 0.9259 0.9568
0.0029 44.0 36784 0.3784 0.9249 0.9282 0.9265 0.9576
0.0019 45.0 37620 0.3895 0.9238 0.9281 0.9259 0.9569
0.0019 46.0 38456 0.3859 0.9238 0.9284 0.9261 0.9572
0.0017 47.0 39292 0.3906 0.9222 0.9277 0.9249 0.9567
0.0016 48.0 40128 0.3933 0.9221 0.9273 0.9247 0.9565
0.0016 49.0 40964 0.3924 0.9224 0.9273 0.9248 0.9565
0.0017 50.0 41800 0.3933 0.9225 0.9275 0.9250 0.9564

Framework versions

  • Transformers 4.34.1
  • Pytorch 2.2.1+cu118
  • Datasets 2.14.6
  • Tokenizers 0.14.1
Downloads last month
3
Inference API
This model can be loaded on Inference API (serverless).

Finetuned from