---
license: mit
base_model: dathi103/bert-job-german-extended
tags:
- generated_from_trainer
model-index:
- name: gerskill-bert-job-extended
  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. -->

# gerskill-bert-job-extended

This model is a fine-tuned version of [dathi103/bert-job-german-extended](https://huggingface.co/dathi103/bert-job-german-extended) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1440
- Hard: {'precision': 0.7093596059113301, 'recall': 0.7933884297520661, 'f1': 0.7490247074122237, 'number': 363}
- Soft: {'precision': 0.7058823529411765, 'recall': 0.7272727272727273, 'f1': 0.7164179104477613, 'number': 66}
- Overall Precision: 0.7089
- Overall Recall: 0.7832
- Overall F1: 0.7442
- Overall Accuracy: 0.9650

## 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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5

### Training results

| Training Loss | Epoch | Step | Validation Loss | Hard                                                                                                     | Soft                                                                                                    | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------------------------------------------------------------------------------------------------------:|:-------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:|
| No log        | 1.0   | 178  | 0.1035          | {'precision': 0.6582278481012658, 'recall': 0.7162534435261708, 'f1': 0.6860158311345645, 'number': 363} | {'precision': 0.6451612903225806, 'recall': 0.6060606060606061, 'f1': 0.625, 'number': 66}              | 0.6565            | 0.6993         | 0.6772     | 0.9597           |
| No log        | 2.0   | 356  | 0.1067          | {'precision': 0.6641414141414141, 'recall': 0.7245179063360881, 'f1': 0.6930171277997365, 'number': 363} | {'precision': 0.676923076923077, 'recall': 0.6666666666666666, 'f1': 0.6717557251908397, 'number': 66}  | 0.6659            | 0.7156         | 0.6899     | 0.9634           |
| 0.1072        | 3.0   | 534  | 0.1204          | {'precision': 0.7079207920792079, 'recall': 0.7878787878787878, 'f1': 0.7457627118644068, 'number': 363} | {'precision': 0.6956521739130435, 'recall': 0.7272727272727273, 'f1': 0.711111111111111, 'number': 66}  | 0.7061            | 0.7786         | 0.7406     | 0.9652           |
| 0.1072        | 4.0   | 712  | 0.1350          | {'precision': 0.7178841309823678, 'recall': 0.7851239669421488, 'f1': 0.7500000000000001, 'number': 363} | {'precision': 0.6956521739130435, 'recall': 0.7272727272727273, 'f1': 0.711111111111111, 'number': 66}  | 0.7146            | 0.7762         | 0.7441     | 0.9644           |
| 0.1072        | 5.0   | 890  | 0.1440          | {'precision': 0.7093596059113301, 'recall': 0.7933884297520661, 'f1': 0.7490247074122237, 'number': 363} | {'precision': 0.7058823529411765, 'recall': 0.7272727272727273, 'f1': 0.7164179104477613, 'number': 66} | 0.7089            | 0.7832         | 0.7442     | 0.9650           |


### Framework versions

- Transformers 4.38.1
- Pytorch 2.1.2+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2