|
--- |
|
license: mit |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: bert-base-german-cased-20000-ner |
|
results: [] |
|
--- |
|
|
|
<!-- 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-base-german-cased-20000-ner |
|
|
|
This model is a fine-tuned version of [bert-base-german-cased](https://huggingface.co/bert-base-german-cased) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0826 |
|
- Precision: 0.8904 |
|
- Recall: 0.8693 |
|
- F1: 0.8797 |
|
- Accuracy: 0.9832 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 32 |
|
- eval_batch_size: 32 |
|
- 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 | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| No log | 0.11 | 64 | 0.0840 | 0.8076 | 0.7842 | 0.7957 | 0.9752 | |
|
| No log | 0.23 | 128 | 0.0787 | 0.8119 | 0.7735 | 0.7922 | 0.9746 | |
|
| No log | 0.34 | 192 | 0.0677 | 0.8264 | 0.8362 | 0.8313 | 0.9794 | |
|
| No log | 0.45 | 256 | 0.0630 | 0.8440 | 0.8125 | 0.8280 | 0.9801 | |
|
| No log | 0.57 | 320 | 0.0664 | 0.8035 | 0.8391 | 0.8209 | 0.9782 | |
|
| No log | 0.68 | 384 | 0.0674 | 0.8850 | 0.8285 | 0.8558 | 0.9819 | |
|
| No log | 0.79 | 448 | 0.0631 | 0.8834 | 0.8598 | 0.8714 | 0.9825 | |
|
| 0.094 | 0.9 | 512 | 0.0572 | 0.8933 | 0.8462 | 0.8691 | 0.9832 | |
|
| 0.094 | 1.02 | 576 | 0.0728 | 0.8520 | 0.8681 | 0.8600 | 0.9795 | |
|
| 0.094 | 1.13 | 640 | 0.0784 | 0.8496 | 0.8717 | 0.8605 | 0.9800 | |
|
| 0.094 | 1.24 | 704 | 0.0721 | 0.8868 | 0.8527 | 0.8695 | 0.9814 | |
|
| 0.094 | 1.36 | 768 | 0.0700 | 0.8755 | 0.8362 | 0.8554 | 0.9808 | |
|
| 0.094 | 1.47 | 832 | 0.0590 | 0.8662 | 0.8610 | 0.8636 | 0.9822 | |
|
| 0.094 | 1.58 | 896 | 0.0615 | 0.8692 | 0.8764 | 0.8728 | 0.9821 | |
|
| 0.094 | 1.7 | 960 | 0.0670 | 0.8812 | 0.8557 | 0.8683 | 0.9826 | |
|
| 0.0413 | 1.81 | 1024 | 0.0623 | 0.9061 | 0.8557 | 0.8802 | 0.9843 | |
|
| 0.0413 | 1.92 | 1088 | 0.0570 | 0.8891 | 0.8770 | 0.8830 | 0.9833 | |
|
| 0.0413 | 2.04 | 1152 | 0.0643 | 0.8859 | 0.8859 | 0.8859 | 0.9831 | |
|
| 0.0413 | 2.15 | 1216 | 0.0705 | 0.8824 | 0.8740 | 0.8782 | 0.9830 | |
|
| 0.0413 | 2.26 | 1280 | 0.0698 | 0.8818 | 0.8557 | 0.8685 | 0.9824 | |
|
| 0.0413 | 2.37 | 1344 | 0.0826 | 0.8904 | 0.8693 | 0.8797 | 0.9832 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.18.0 |
|
- Pytorch 1.9.0+cu111 |
|
- Datasets 2.1.0 |
|
- Tokenizers 0.12.1 |
|
|
