Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,95 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
license: apache-2.0
|
| 5 |
+
base_model: bert-base-multilingual-cased
|
| 6 |
+
tags:
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- tmnam20/VieGLUE
|
| 10 |
+
metrics:
|
| 11 |
+
- accuracy
|
| 12 |
+
model-index:
|
| 13 |
+
- name: bert-base-multilingual-cased-qnli-1
|
| 14 |
+
results:
|
| 15 |
+
- task:
|
| 16 |
+
name: Text Classification
|
| 17 |
+
type: text-classification
|
| 18 |
+
dataset:
|
| 19 |
+
name: tmnam20/VieGLUE/QNLI
|
| 20 |
+
type: tmnam20/VieGLUE
|
| 21 |
+
config: qnli
|
| 22 |
+
split: validation
|
| 23 |
+
args: qnli
|
| 24 |
+
metrics:
|
| 25 |
+
- name: Accuracy
|
| 26 |
+
type: accuracy
|
| 27 |
+
value: 0.885227896760022
|
| 28 |
+
---
|
| 29 |
+
|
| 30 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 31 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 32 |
+
|
| 33 |
+
# bert-base-multilingual-cased-qnli-1
|
| 34 |
+
|
| 35 |
+
This model is a fine-tuned version of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) on the tmnam20/VieGLUE/QNLI dataset.
|
| 36 |
+
It achieves the following results on the evaluation set:
|
| 37 |
+
- Loss: 0.3278
|
| 38 |
+
- Accuracy: 0.8852
|
| 39 |
+
|
| 40 |
+
## Model description
|
| 41 |
+
|
| 42 |
+
More information needed
|
| 43 |
+
|
| 44 |
+
## Intended uses & limitations
|
| 45 |
+
|
| 46 |
+
More information needed
|
| 47 |
+
|
| 48 |
+
## Training and evaluation data
|
| 49 |
+
|
| 50 |
+
More information needed
|
| 51 |
+
|
| 52 |
+
## Training procedure
|
| 53 |
+
|
| 54 |
+
### Training hyperparameters
|
| 55 |
+
|
| 56 |
+
The following hyperparameters were used during training:
|
| 57 |
+
- learning_rate: 2e-05
|
| 58 |
+
- train_batch_size: 32
|
| 59 |
+
- eval_batch_size: 16
|
| 60 |
+
- seed: 1
|
| 61 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 62 |
+
- lr_scheduler_type: linear
|
| 63 |
+
- num_epochs: 3.0
|
| 64 |
+
|
| 65 |
+
### Training results
|
| 66 |
+
|
| 67 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 68 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 69 |
+
| 0.3938 | 0.15 | 500 | 0.3494 | 0.8495 |
|
| 70 |
+
| 0.3712 | 0.31 | 1000 | 0.3266 | 0.8570 |
|
| 71 |
+
| 0.3837 | 0.46 | 1500 | 0.3174 | 0.8655 |
|
| 72 |
+
| 0.3466 | 0.61 | 2000 | 0.2957 | 0.8785 |
|
| 73 |
+
| 0.3084 | 0.76 | 2500 | 0.3093 | 0.8715 |
|
| 74 |
+
| 0.322 | 0.92 | 3000 | 0.2950 | 0.8731 |
|
| 75 |
+
| 0.273 | 1.07 | 3500 | 0.2872 | 0.8834 |
|
| 76 |
+
| 0.2628 | 1.22 | 4000 | 0.3110 | 0.8794 |
|
| 77 |
+
| 0.2732 | 1.37 | 4500 | 0.2910 | 0.8797 |
|
| 78 |
+
| 0.2592 | 1.53 | 5000 | 0.2855 | 0.8849 |
|
| 79 |
+
| 0.241 | 1.68 | 5500 | 0.2974 | 0.8861 |
|
| 80 |
+
| 0.2256 | 1.83 | 6000 | 0.2914 | 0.8850 |
|
| 81 |
+
| 0.2402 | 1.99 | 6500 | 0.2759 | 0.8883 |
|
| 82 |
+
| 0.1958 | 2.14 | 7000 | 0.3080 | 0.8880 |
|
| 83 |
+
| 0.1684 | 2.29 | 7500 | 0.3190 | 0.8847 |
|
| 84 |
+
| 0.1472 | 2.44 | 8000 | 0.3305 | 0.8871 |
|
| 85 |
+
| 0.1601 | 2.6 | 8500 | 0.3298 | 0.8836 |
|
| 86 |
+
| 0.1857 | 2.75 | 9000 | 0.3274 | 0.8847 |
|
| 87 |
+
| 0.1667 | 2.9 | 9500 | 0.3256 | 0.8841 |
|
| 88 |
+
|
| 89 |
+
|
| 90 |
+
### Framework versions
|
| 91 |
+
|
| 92 |
+
- Transformers 4.35.2
|
| 93 |
+
- Pytorch 2.2.0.dev20231203+cu121
|
| 94 |
+
- Datasets 2.15.0
|
| 95 |
+
- Tokenizers 0.15.0
|