Downloads · 30 days
17
1% of all-time downloads
bert-german-ner
lunesco/bert-german-ner
bert-german-ner is a token classification model from lunesco. Use it when you need labels on individual words, such as names. 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
1.4K
Public
Parameters
109M
1.8 GB on disk
Likes
1
Public
Repo makeup
Click a slice to open those files.
.bin438 MB · 50%
Parameter types
How the weights are stored.
F32109M · 100%
Try a prompt
Base models
Datasets
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. -->
bert-german-ner
This model is a fine-tuned version of dbmdz/bert-base-german-cased on the conll2003 dataset. It achieves the following results on the evaluation set:
- Loss: 0.3196
- Precision: 0.8334
- Recall: 0.8620
- F1: 0.8474
- Accuracy: 0.9292
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: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 300 | 0.3617 | 0.7310 | 0.7733 | 0.7516 | 0.8908 |
| 0.5428 | 2.0 | 600 | 0.2897 | 0.7789 | 0.8395 | 0.8081 | 0.9132 |
| 0.5428 | 3.0 | 900 | 0.2805 | 0.8147 | 0.8465 | 0.8303 | 0.9221 |
| 0.2019 | 4.0 | 1200 | 0.2816 | 0.8259 | 0.8498 | 0.8377 | 0.9260 |
| 0.1215 | 5.0 | 1500 | 0.2942 | 0.8332 | 0.8599 | 0.8463 | 0.9285 |
| 0.1215 | 6.0 | 1800 | 0.3053 | 0.8293 | 0.8619 | 0.8452 | 0.9287 |
| 0.0814 | 7.0 | 2100 | 0.3190 | 0.8249 | 0.8634 | 0.8437 | 0.9267 |
| 0.0814 | 8.0 | 2400 | 0.3196 | 0.8334 | 0.8620 | 0.8474 | 0.9292 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.13.1+cu116
- Datasets 2.9.0
- Tokenizers 0.13.2
- Task
- Token Classification
- Library
- transformers
- Type
- bert
- License
- mit
- Created
- Jan 30, 2023
- Updated
- Dec 20, 2023