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

This model is a fine-tuned version of [dathi103/bert-job-german](https://huggingface.co/dathi103/bert-job-german) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1171
- Hard: {'precision': 0.7529644268774703, 'recall': 0.8355263157894737, 'f1': 0.7920997920997921, 'number': 456}
- Soft: {'precision': 0.7906976744186046, 'recall': 0.8292682926829268, 'f1': 0.8095238095238095, 'number': 82}
- Overall Precision: 0.7584
- Overall Recall: 0.8346
- Overall F1: 0.7947
- Overall Accuracy: 0.9675

## 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  | 0.1177          | {'precision': 0.6027397260273972, 'recall': 0.7719298245614035, 'f1': 0.676923076923077, 'number': 456}  | {'precision': 0.6629213483146067, 'recall': 0.7195121951219512, 'f1': 0.6900584795321638, 'number': 82} | 0.6107            | 0.7639         | 0.6788     | 0.9524           |
| No log        | 2.0   | 356  | 0.0978          | {'precision': 0.7474541751527495, 'recall': 0.8048245614035088, 'f1': 0.775079197465681, 'number': 456}  | {'precision': 0.7058823529411765, 'recall': 0.7317073170731707, 'f1': 0.718562874251497, 'number': 82}  | 0.7413            | 0.7937         | 0.7666     | 0.9620           |
| 0.1344        | 3.0   | 534  | 0.1022          | {'precision': 0.7242718446601941, 'recall': 0.8179824561403509, 'f1': 0.768280123583934, 'number': 456}  | {'precision': 0.735632183908046, 'recall': 0.7804878048780488, 'f1': 0.757396449704142, 'number': 82}   | 0.7259            | 0.8123         | 0.7667     | 0.9632           |
| 0.1344        | 4.0   | 712  | 0.1133          | {'precision': 0.762278978388998, 'recall': 0.8508771929824561, 'f1': 0.8041450777202073, 'number': 456}  | {'precision': 0.7555555555555555, 'recall': 0.8292682926829268, 'f1': 0.7906976744186047, 'number': 82} | 0.7613            | 0.8476         | 0.8021     | 0.9665           |
| 0.1344        | 5.0   | 890  | 0.1171          | {'precision': 0.7529644268774703, 'recall': 0.8355263157894737, 'f1': 0.7920997920997921, 'number': 456} | {'precision': 0.7906976744186046, 'recall': 0.8292682926829268, 'f1': 0.8095238095238095, 'number': 82} | 0.7584            | 0.8346         | 0.7947     | 0.9675           |


### Framework versions

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