Downloads · 30 days
20
10% of all-time downloads
variant_tapt_freeze_llrd_LR_5e
Mardiyyah/variant_tapt_freeze_llrd_LR_5e
variant_tapt_freeze_llrd_LR_5e is a token classification model from Mardiyyah. Use it when you need labels on individual words, such as names. 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
195
Public
Parameters
110M
13.2 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors440 MB · 100%
At a glance
- Task
- Token Classification
- Library
- transformers
- License
- apache-2.0
- Model type
- bert
- Access
- Public
- Created
- Jan 6, 2026
- Updated
- Jan 6, 2026
- SHA
- be796a2f
Try a prompt
Base models
Spaces
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. -->
variant_tapt_freeze_llrd_LR_5e
This model is a fine-tuned version of Mardiyyah/variant-tapt_freeze_llrd-LR_5e-05 on the OTAR3088/PDBE_IOB_all dataset. It achieves the following results on the evaluation set:
- Loss: 0.2349
- Precision: 0.9118
- Recall: 0.9193
- F1: 0.9155
- Accuracy: 0.9617
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: 16
- eval_batch_size: 16
- seed: 3407
- 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
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 1.9616 | 1.0 | 352 | 0.8090 | 0.4440 | 0.4431 | 0.4435 | 0.7943 |
| 0.5637 | 2.0 | 704 | 0.3524 | 0.7245 | 0.7551 | 0.7395 | 0.9077 |
| 0.2766 | 3.0 | 1056 | 0.2292 | 0.8258 | 0.8408 | 0.8332 | 0.9383 |
| 0.1736 | 4.0 | 1408 | 0.2102 | 0.8722 | 0.8553 | 0.8637 | 0.9463 |
| 0.1209 | 5.0 | 1760 | 0.1951 | 0.8693 | 0.8861 | 0.8776 | 0.9506 |
| 0.0846 | 6.0 | 2112 | 0.1981 | 0.8914 | 0.8908 | 0.8911 | 0.9544 |
| 0.0641 | 7.0 | 2464 | 0.1917 | 0.8956 | 0.9030 | 0.8993 | 0.9560 |
| 0.0485 | 8.0 | 2816 | 0.2001 | 0.8958 | 0.9096 | 0.9027 | 0.9575 |
| 0.0374 | 9.0 | 3168 | 0.1962 | 0.8938 | 0.9123 | 0.9030 | 0.9577 |
| 0.0299 | 10.0 | 3520 | 0.2070 | 0.8966 | 0.9160 | 0.9062 | 0.9590 |
| 0.0238 | 11.0 | 3872 | 0.2104 | 0.9030 | 0.9171 | 0.9100 | 0.9591 |
| 0.0195 | 12.0 | 4224 | 0.2170 | 0.8996 | 0.9169 | 0.9082 | 0.9593 |
| 0.0158 | 13.0 | 4576 | 0.2184 | 0.9051 | 0.9183 | 0.9116 | 0.9608 |
| 0.013 | 14.0 | 4928 | 0.2255 | 0.9017 | 0.9211 | 0.9113 | 0.9602 |
| 0.011 | 15.0 | 5280 | 0.2366 | 0.9037 | 0.9190 | 0.9113 | 0.9600 |
| 0.0101 | 16.0 | 5632 | 0.2317 | 0.9092 | 0.9215 | 0.9153 | 0.9613 |
| 0.0083 | 17.0 | 5984 | 0.2344 | 0.9118 | 0.9193 | 0.9155 | 0.9617 |
| 0.007 | 18.0 | 6336 | 0.2400 | 0.9091 | 0.9214 | 0.9152 | 0.9615 |
| 0.0067 | 19.0 | 6688 | 0.2395 | 0.9075 | 0.9213 | 0.9144 | 0.9611 |
| 0.0063 | 20.0 | 7040 | 0.2420 | 0.9084 | 0.9208 | 0.9145 | 0.9611 |
Framework versions
- Transformers 4.48.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.2
- Tokenizers 0.21.0
- Task
- Token Classification
- Library
- transformers
- Type
- bert
- License
- apache-2.0
- Languages
- en
- Created
- Jan 6, 2026
- Updated
- Jan 6, 2026