Downloads · 30 days
20
2% of all-time downloads
bertweet-base-SNS_BRANDS_50k
haji2438/bertweet-base-SNS_BRANDS_50k
bertweet-base-SNS_BRANDS_50k is a fill-mask model from haji2438. Use it when you need the model to fill a missing word. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
All-time downloads
1K
Public
Repo size
4.9 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin540 MB · 100%
At a glance
- Task
- Fill-Mask
- Library
- transformers
- Model type
- roberta
- Access
- Public
- Created
- Mar 2, 2022
- Updated
- Jan 24, 2022
- SHA
- 4a908f42
Try a prompt
Model card
From the Hugging Face model README
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
bertweet-base-SNS_BRANDS_50k
This model is a fine-tuned version of vinai/bertweet-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0490
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.0787 | 1.0 | 1465 | 0.0751 |
| 0.0662 | 2.0 | 2930 | 0.0628 |
| 0.053 | 3.0 | 4395 | 0.0531 |
| 0.0452 | 4.0 | 5860 | 0.0490 |
Framework versions
- Transformers 4.15.0
- Pytorch 1.10.0+cu111
- Datasets 1.18.0
- Tokenizers 0.10.3
- Task
- Fill-Mask
- Library
- transformers
- Type
- roberta
- Created
- Mar 2, 2022
- Updated
- Jan 24, 2022