Downloads · 30 days
10
2% of all-time downloads
albert-xxlarge-v2-finetuned-Poems
Hausax/albert-xxlarge-v2-finetuned-Poems
albert-xxlarge-v2-finetuned-Poems is a fill-mask model from Hausax. Use it when you need the model to fill a missing word. It is set up for transformers. The card lists the license as apache-2.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
619
Public
Repo size
4.1 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin825 MB · 100%
At a glance
- Task
- Fill-Mask
- Library
- transformers
- License
- apache-2.0
- Model type
- albert
- Access
- Public
- Created
- Jun 19, 2022
- Updated
- Jun 20, 2022
- SHA
- a2095801
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. -->
albert-xxlarge-v2-finetuned-Poems
This model is a fine-tuned version of albert-xxlarge-v2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.1923
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-07
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.482 | 1.0 | 19375 | 2.2959 |
| 2.258 | 2.0 | 38750 | 2.2357 |
| 2.2146 | 3.0 | 58125 | 2.2085 |
| 2.1975 | 4.0 | 77500 | 2.1929 |
| 2.1893 | 5.0 | 96875 | 2.1863 |
Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1
- Task
- Fill-Mask
- Library
- transformers
- Type
- albert
- License
- apache-2.0
- Created
- Jun 19, 2022
- Updated
- Jun 20, 2022