Edit model card

not-ner-v2_16batch

This model is a fine-tuned version of dccuchile/tulio-chilean-spanish-bert on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1124
  • Accuracy: 0.9662
  • Precision: 0.9671
  • Recall: 0.9662
  • F1: 0.9665

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: 5e-05
  • train_batch_size: 16
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss Accuracy Precision Recall F1
0.0906 0.0798 250 0.1124 0.9662 0.9671 0.9662 0.9665
0.1008 0.1596 500 0.1416 0.9654 0.9652 0.9654 0.9653
0.0918 0.2394 750 0.1361 0.9662 0.9657 0.9662 0.9659
0.1077 0.3192 1000 0.1353 0.9636 0.9626 0.9636 0.9627
0.0863 0.3990 1250 0.1457 0.9590 0.9624 0.9590 0.9601
0.1048 0.4788 1500 0.1340 0.9653 0.9645 0.9653 0.9648
0.1066 0.5586 1750 0.1490 0.9631 0.9621 0.9631 0.9624
0.107 0.6384 2000 0.1479 0.9638 0.9641 0.9638 0.9639
0.1274 0.7182 2250 0.1331 0.9636 0.9634 0.9636 0.9635
0.1381 0.7980 2500 0.1577 0.9626 0.9616 0.9626 0.9618
0.1111 0.8778 2750 0.1353 0.9625 0.9637 0.9625 0.9630
0.132 0.9575 3000 0.1785 0.9592 0.9578 0.9592 0.9579
0.1464 1.0373 3250 0.1712 0.9610 0.9598 0.9610 0.9600
0.1279 1.1171 3500 0.1514 0.9630 0.9622 0.9630 0.9625
0.0947 1.1969 3750 0.1554 0.9625 0.9629 0.9625 0.9627
0.0958 1.2767 4000 0.1746 0.9617 0.9606 0.9617 0.9607
0.1212 1.3565 4250 0.1494 0.9628 0.9620 0.9628 0.9622
0.1097 1.4363 4500 0.1557 0.9635 0.9627 0.9635 0.9630
0.1288 1.5161 4750 0.1499 0.9642 0.9633 0.9642 0.9635
0.1043 1.5959 5000 0.1488 0.9614 0.9619 0.9614 0.9617
0.1016 1.6757 5250 0.1521 0.9616 0.9611 0.9616 0.9613
0.1073 1.7555 5500 0.1455 0.9618 0.9621 0.9618 0.9619
0.1264 1.8353 5750 0.2369 0.9143 0.9407 0.9143 0.9218
0.1211 1.9151 6000 0.1524 0.9613 0.9606 0.9613 0.9609
0.1071 1.9949 6250 0.1327 0.9574 0.9565 0.9574 0.9548
0.0915 2.0747 6500 0.1325 0.9640 0.9638 0.9640 0.9639
0.084 2.1545 6750 0.1321 0.9647 0.9646 0.9647 0.9646
0.0922 2.2343 7000 0.1486 0.9644 0.9635 0.9644 0.9637
0.0877 2.3141 7250 0.1281 0.9643 0.9648 0.9643 0.9645
0.0828 2.3939 7500 0.1425 0.9575 0.9591 0.9575 0.9582
0.0859 2.4737 7750 0.1283 0.9670 0.9665 0.9670 0.9667
0.0866 2.5535 8000 0.1189 0.9677 0.9671 0.9677 0.9673
0.0681 2.6333 8250 0.1195 0.9689 0.9682 0.9689 0.9683
0.0462 2.7131 8500 0.1408 0.9678 0.9670 0.9678 0.9670
0.0672 2.7929 8750 0.1233 0.9690 0.9684 0.9690 0.9686
0.0578 2.8726 9000 0.1232 0.9694 0.9689 0.9694 0.9691
0.0591 2.9524 9250 0.1219 0.9694 0.9689 0.9694 0.9690

Framework versions

  • Transformers 4.42.4
  • Pytorch 2.3.1+cu121
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
0
Safetensors
Model size
110M params
Tensor type
F32
·
Inference API
Unable to determine this model's library. Check the docs .

Model tree for lmaccarini/not-ner-v2_16batch

Finetuned
(9)
this model