ViGLUE
Collection
A collection to store all the artifacts of the paper: ViGLUE: A Vietnamese General Language Understanding Benchmark and Analysis of Vietnamese LMs. • 145 items • Updated
bert-base-multilingual-cased-mnli-100
This model is a fine-tuned version of bert-base-multilingual-cased on the tmnam20/VieGLUE/MNLI dataset. It achieves the following results on the evaluation set:
- Loss: 0.5343
- Accuracy: 0.8063
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: 16
- seed: 100
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.62 | 0.41 | 5000 | 0.6193 | 0.7459 |
| 0.5923 | 0.81 | 10000 | 0.5911 | 0.7610 |
| 0.5136 | 1.22 | 15000 | 0.5670 | 0.7808 |
| 0.4927 | 1.63 | 20000 | 0.5558 | 0.7852 |
| 0.4425 | 2.04 | 25000 | 0.5809 | 0.7844 |
| 0.4301 | 2.44 | 30000 | 0.5546 | 0.7940 |
| 0.4017 | 2.85 | 35000 | 0.5565 | 0.7963 |
Framework versions
- Transformers 4.36.0
- Pytorch 2.1.0+cu121
- Datasets 2.15.0
- Tokenizers 0.15.0
- Downloads last month
- 5
Model tree for tmnam20/bert-base-multilingual-cased-mnli-100
Base model
google-bert/bert-base-multilingual-casedDataset used to train tmnam20/bert-base-multilingual-cased-mnli-100
Collection including tmnam20/bert-base-multilingual-cased-mnli-100
Evaluation results
- Accuracy on tmnam20/VieGLUE/MNLIself-reported0.806