haryoaw's picture
Initial Commit
4674b55 verified
|
raw
history blame
2.34 kB
metadata
license: mit
base_model: microsoft/mdeberta-v3-base
tags:
  - generated_from_trainer
datasets:
  - massive
metrics:
  - accuracy
  - f1
model-index:
  - name: scenario-MDBT-TCR_data-cl-massive_all_1_1
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: massive
          type: massive
          config: all_1.1
          split: validation
          args: all_1.1
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8000036435845584
          - name: F1
            type: f1
            value: 0.758305292124411

scenario-MDBT-TCR_data-cl-massive_all_1_1

This model is a fine-tuned version of microsoft/mdeberta-v3-base on the massive dataset. It achieves the following results on the evaluation set:

  • Loss: 1.3801
  • Accuracy: 0.8000
  • F1: 0.7583

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: 32
  • eval_batch_size: 64
  • seed: 66
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 5

Training results

Training Loss Epoch Step Validation Loss Accuracy F1
0.489 0.56 5000 0.9200 0.7871 0.7332
0.2753 1.11 10000 1.0539 0.7885 0.7364
0.2348 1.67 15000 1.0362 0.7891 0.7431
0.1648 2.22 20000 1.1925 0.7867 0.7535
0.1481 2.78 25000 1.1608 0.7920 0.7513
0.1074 3.33 30000 1.3151 0.7966 0.7513
0.0942 3.89 35000 1.3274 0.7952 0.7523
0.0684 4.45 40000 1.3801 0.8000 0.7583

Framework versions

  • Transformers 4.33.3
  • Pytorch 2.1.1+cu121
  • Datasets 2.14.5
  • Tokenizers 0.13.3