README.md

metadata

license: apache-2.0
tags:
  - generated_from_keras_callback
model-index:
  - name: dosai/berts-finetuned-squad
    results: []

dosai/berts-finetuned-squad

This model is a fine-tuned version of bert-base-multilingual-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.0077
• Epoch: 29

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:

• optimizer: {'inner_optimizer': {'class_name': 'AdamWeightDecay', 'config': {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 1020, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.001}}, 'dynamic': True, 'initial_scale': 32768.0, 'dynamic_growth_steps': 2000}
• training_precision: mixed_float16

Training results

Train Loss	Epoch
3.6463	0
1.7361	1
1.2498	2
0.8598	3
0.7020	4
0.3977	5
0.2314	6
0.1092	7
0.0666	8
0.0521	9
0.0412	10
0.0354	11
0.0269	12
0.0177	13
0.0184	14
0.0369	15
0.0198	16
0.0169	17
0.0132	18
0.0091	19
0.0080	20
0.0093	21
0.0065	22
0.0064	23
0.0108	24
0.0080	25
0.0070	26
0.0071	27
0.0061	28
0.0077	29

Framework versions

• Transformers 4.29.2
• TensorFlow 2.12.0
• Datasets 2.12.0
• Tokenizers 0.13.3