stulcrad's picture
Model save
19ecf99 verified
metadata
base_model: UWB-AIR/Czert-B-base-cased
tags:
  - generated_from_trainer
datasets:
  - cnec
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: CNEC_2_0_Czert-B-base-cased
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: cnec
          type: cnec
          config: default
          split: validation
          args: default
        metrics:
          - name: Precision
            type: precision
            value: 0.8093464273620048
          - name: Recall
            type: recall
            value: 0.8547925608011445
          - name: F1
            type: f1
            value: 0.8314489476430683
          - name: Accuracy
            type: accuracy
            value: 0.9446311123820418

CNEC_2_0_Czert-B-base-cased

This model is a fine-tuned version of UWB-AIR/Czert-B-base-cased on the cnec dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3352
  • Precision: 0.8093
  • Recall: 0.8548
  • F1: 0.8314
  • Accuracy: 0.9446

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: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 25

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
0.5496 2.22 500 0.2782 0.7301 0.7750 0.7519 0.9275
0.2133 4.44 1000 0.2487 0.7811 0.8219 0.8010 0.9399
0.144 6.67 1500 0.2580 0.7737 0.8290 0.8004 0.9396
0.1029 8.89 2000 0.2576 0.7997 0.8480 0.8231 0.9446
0.0776 11.11 2500 0.2849 0.7990 0.8516 0.8244 0.9444
0.0601 13.33 3000 0.2971 0.8021 0.8523 0.8264 0.9450
0.0494 15.56 3500 0.3077 0.8014 0.8473 0.8237 0.9440
0.0408 17.78 4000 0.3145 0.8131 0.8555 0.8337 0.9448
0.0353 20.0 4500 0.3260 0.8097 0.8569 0.8327 0.9445
0.0311 22.22 5000 0.3356 0.8076 0.8541 0.8302 0.9441
0.0281 24.44 5500 0.3352 0.8093 0.8548 0.8314 0.9446

Framework versions

  • Transformers 4.36.2
  • Pytorch 2.1.2+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.0