Downloads · 30 days
19
4% of all-time downloads
mlm-heb-medical
cp500/mlm-heb-medical
mlm-heb-medical is a fill-mask model from cp500. 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
493
Public
Repo size
96.7 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin738 MB · 99%
At a glance
- Task
- Fill-Mask
- Library
- transformers
- Model type
- bert
- Access
- Public
- Created
- Oct 31, 2023
- Updated
- Nov 2, 2023
- SHA
- 58295742
Try a prompt
Base models
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. -->
mlm-heb-medical
This model is a fine-tuned version of imvladikon/alephbertgimmel-base-512 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.7757
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
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.4809 | 1.0 | 5304 | 1.3390 |
| 1.2434 | 2.0 | 10608 | 1.1131 |
| 1.105 | 3.0 | 15912 | 0.9920 |
| 1.0178 | 4.0 | 21216 | 0.9150 |
| 0.961 | 5.0 | 26520 | 0.8706 |
| 0.9326 | 6.0 | 31824 | 0.8399 |
| 0.9119 | 7.0 | 37128 | 0.8179 |
| 0.8716 | 8.0 | 42432 | 0.7860 |
| 0.8644 | 9.0 | 47736 | 0.7903 |
| 0.8666 | 10.0 | 53040 | 0.7676 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.0.1
- Datasets 2.14.6
- Tokenizers 0.14.1
- Task
- Fill-Mask
- Library
- transformers
- Type
- bert
- Created
- Oct 31, 2023
- Updated
- Nov 2, 2023