README.md

metadata

library_name: transformers
language:
  - en
base_model: Hartunka/tiny_bert_km_5_v2
tags:
  - generated_from_trainer
datasets:
  - glue
metrics:
  - accuracy
  - f1
model-index:
  - name: tiny_bert_km_5_v2_qqp
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: GLUE QQP
          type: glue
          args: qqp
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8236210734603018
          - name: F1
            type: f1
            value: 0.7516109930683758

tiny_bert_km_5_v2_qqp

This model is a fine-tuned version of Hartunka/tiny_bert_km_5_v2 on the GLUE QQP dataset. It achieves the following results on the evaluation set:

• Loss: 0.4044
• Accuracy: 0.8236
• F1: 0.7516
• Combined Score: 0.7876

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: 256
• eval_batch_size: 256
• seed: 10
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Combined Score
0.4976	1.0	1422	0.4732	0.7705	0.6348	0.7027
0.4095	2.0	2844	0.4118	0.8045	0.7307	0.7676
0.3541	3.0	4266	0.4059	0.8161	0.7445	0.7803
0.3088	4.0	5688	0.4044	0.8236	0.7516	0.7876
0.2715	5.0	7110	0.4359	0.8229	0.7362	0.7796
0.2395	6.0	8532	0.4361	0.8292	0.7621	0.7956
0.2125	7.0	9954	0.4422	0.8311	0.7741	0.8026
0.1876	8.0	11376	0.4813	0.8353	0.7755	0.8054
0.1677	9.0	12798	0.5044	0.8330	0.7809	0.8070

Framework versions

• Transformers 4.50.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.21.1