metadata

license: apache-2.0
base_model: google/bert_uncased_L-4_H-256_A-4
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: tinybert-TG-HS-HX-parentpretrained
    results: []

tinybert-TG-HS-HX-parentpretrained

This model is a fine-tuned version of google/bert_uncased_L-4_H-256_A-4 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.1937
Accuracy: 0.8230

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: 5.2898091511494136e-05
train_batch_size: 128
eval_batch_size: 128
seed: 33
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 7

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.1819	1.0	197	0.1923	0.8227
0.1791	2.0	394	0.1922	0.8223
0.1772	3.0	591	0.1950	0.8143
0.1761	4.0	788	0.1932	0.8239
0.1756	5.0	985	0.1932	0.8234
0.1752	6.0	1182	0.1939	0.8242
0.1759	7.0	1379	0.1937	0.8230

Framework versions

Transformers 4.36.0.dev0
Pytorch 2.1.1
Datasets 2.15.0
Tokenizers 0.15.0