Downloads · 30 days
69
36% of all-time downloads
adaptation_diacritized
lazymonster/adaptation_diacritized
adaptation_diacritized is a text classification model from lazymonster. Use it when you need a label for a piece of text. It is set up for transformers. The card lists the license as mit.
should probably proofread and complete it, then remove this comment. --
All-time downloads
191
Public
Parameters
16.1M
1.5 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors64.5 MB · 99%
At a glance
- Task
- Text Classification
- Library
- transformers
- License
- mit
- Model type
- roberta
- Access
- Public
- Created
- Dec 2, 2025
- Updated
- Dec 2, 2025
- SHA
- 52c949b4
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. -->
adaptation_diacritized
This model is a fine-tuned version of lazymonster/15k-combined-diacritized-scratch on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.0240
- Accuracy: 0.6734
- F1 Macro: 0.6510
- Composite Score: 0.6510
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 20
- mixed_precision_training: Native AMP
- label_smoothing_factor: 0.1
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | Composite Score |
|---|---|---|---|---|---|---|
| No log | 1.0 | 267 | 0.9467 | 0.6662 | 0.6397 | 0.6397 |
| 0.5963 | 2.0 | 534 | 0.9685 | 0.6504 | 0.6373 | 0.6373 |
| 0.5963 | 3.0 | 801 | 0.9401 | 0.6686 | 0.6457 | 0.6457 |
| 0.5689 | 4.0 | 1068 | 0.9578 | 0.6671 | 0.6433 | 0.6433 |
| 0.5689 | 5.0 | 1335 | 0.9699 | 0.6681 | 0.6452 | 0.6452 |
| 0.5488 | 6.0 | 1602 | 0.9692 | 0.6758 | 0.6484 | 0.6484 |
| 0.5488 | 7.0 | 1869 | 0.9762 | 0.6715 | 0.6513 | 0.6513 |
| 0.5263 | 8.0 | 2136 | 0.9864 | 0.6786 | 0.6530 | 0.6530 |
| 0.5263 | 9.0 | 2403 | 0.9853 | 0.6777 | 0.6523 | 0.6523 |
| 0.5072 | 10.0 | 2670 | 0.9843 | 0.6844 | 0.6573 | 0.6573 |
| 0.5072 | 11.0 | 2937 | 1.0037 | 0.6782 | 0.6553 | 0.6553 |
| 0.4946 | 12.0 | 3204 | 0.9960 | 0.6767 | 0.6548 | 0.6548 |
| 0.4946 | 13.0 | 3471 | 1.0123 | 0.6810 | 0.6572 | 0.6572 |
| 0.488 | 14.0 | 3738 | 1.0240 | 0.6734 | 0.6510 | 0.6510 |
Framework versions
- Transformers 4.53.3
- Pytorch 2.6.0+cu124
- Datasets 4.4.1
- Tokenizers 0.21.2
- Task
- Text Classification
- Library
- transformers
- Type
- roberta
- License
- mit
- Created
- Dec 2, 2025
- Updated
- Dec 2, 2025