update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
tags:
|
| 4 |
+
- question-answering
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
model-index:
|
| 7 |
+
- name: roberta_qa_japanese
|
| 8 |
+
results: []
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 13 |
+
|
| 14 |
+
# roberta_qa_japanese
|
| 15 |
+
|
| 16 |
+
This model is a fine-tuned version of [rinna/japanese-roberta-base](https://huggingface.co/rinna/japanese-roberta-base) on the None dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 0.0516
|
| 19 |
+
|
| 20 |
+
## Model description
|
| 21 |
+
|
| 22 |
+
More information needed
|
| 23 |
+
|
| 24 |
+
## Intended uses & limitations
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Training and evaluation data
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training procedure
|
| 33 |
+
|
| 34 |
+
### Training hyperparameters
|
| 35 |
+
|
| 36 |
+
The following hyperparameters were used during training:
|
| 37 |
+
- learning_rate: 7e-05
|
| 38 |
+
- train_batch_size: 2
|
| 39 |
+
- eval_batch_size: 1
|
| 40 |
+
- seed: 42
|
| 41 |
+
- gradient_accumulation_steps: 16
|
| 42 |
+
- total_train_batch_size: 32
|
| 43 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
+
- lr_scheduler_type: linear
|
| 45 |
+
- lr_scheduler_warmup_steps: 100
|
| 46 |
+
- num_epochs: 3
|
| 47 |
+
|
| 48 |
+
### Training results
|
| 49 |
+
|
| 50 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 51 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 52 |
+
| 2.1293 | 0.13 | 150 | 1.0311 |
|
| 53 |
+
| 1.1965 | 0.26 | 300 | 0.6723 |
|
| 54 |
+
| 1.022 | 0.39 | 450 | 0.4838 |
|
| 55 |
+
| 0.9594 | 0.53 | 600 | 0.5174 |
|
| 56 |
+
| 0.9187 | 0.66 | 750 | 0.4671 |
|
| 57 |
+
| 0.8229 | 0.79 | 900 | 0.4650 |
|
| 58 |
+
| 0.71 | 0.92 | 1050 | 0.2648 |
|
| 59 |
+
| 0.5436 | 1.05 | 1200 | 0.2665 |
|
| 60 |
+
| 0.5045 | 1.19 | 1350 | 0.2686 |
|
| 61 |
+
| 0.5025 | 1.32 | 1500 | 0.2082 |
|
| 62 |
+
| 0.5213 | 1.45 | 1650 | 0.1715 |
|
| 63 |
+
| 0.4648 | 1.58 | 1800 | 0.1563 |
|
| 64 |
+
| 0.4698 | 1.71 | 1950 | 0.1488 |
|
| 65 |
+
| 0.4823 | 1.84 | 2100 | 0.1050 |
|
| 66 |
+
| 0.4482 | 1.97 | 2250 | 0.0821 |
|
| 67 |
+
| 0.2755 | 2.11 | 2400 | 0.0898 |
|
| 68 |
+
| 0.2834 | 2.24 | 2550 | 0.0964 |
|
| 69 |
+
| 0.2525 | 2.37 | 2700 | 0.0533 |
|
| 70 |
+
| 0.2606 | 2.5 | 2850 | 0.0561 |
|
| 71 |
+
| 0.2467 | 2.63 | 3000 | 0.0601 |
|
| 72 |
+
| 0.2799 | 2.77 | 3150 | 0.0562 |
|
| 73 |
+
| 0.2497 | 2.9 | 3300 | 0.0516 |
|
| 74 |
+
|
| 75 |
+
|
| 76 |
+
### Framework versions
|
| 77 |
+
|
| 78 |
+
- Transformers 4.23.1
|
| 79 |
+
- Pytorch 1.12.1+cu102
|
| 80 |
+
- Datasets 2.6.1
|
| 81 |
+
- Tokenizers 0.13.1
|