metadata
license: mit
base_model: dathi103/gbert-job-extended
tags:
- generated_from_trainer
model-index:
- name: gerskill-gbert-job-extended
results: []
gerskill-gbert-job-extended
This model is a fine-tuned version of dathi103/gbert-job-extended on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1217
- Hard: {'precision': 0.7340153452685422, 'recall': 0.790633608815427, 'f1': 0.7612732095490715, 'number': 363}
- Soft: {'precision': 0.6911764705882353, 'recall': 0.7121212121212122, 'f1': 0.7014925373134329, 'number': 66}
- Overall Precision: 0.7277
- Overall Recall: 0.7786
- Overall F1: 0.7523
- Overall Accuracy: 0.9661
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 | 0.1108 | {'precision': 0.6256038647342995, 'recall': 0.7134986225895317, 'f1': 0.6666666666666667, 'number': 363} | {'precision': 0.5606060606060606, 'recall': 0.5606060606060606, 'f1': 0.5606060606060606, 'number': 66} | 0.6167 | 0.6900 | 0.6513 | 0.9593 |
| No log | 2.0 | 356 | 0.1027 | {'precision': 0.6860759493670886, 'recall': 0.7465564738292011, 'f1': 0.7150395778364115, 'number': 363} | {'precision': 0.7096774193548387, 'recall': 0.6666666666666666, 'f1': 0.6875, 'number': 66} | 0.6893 | 0.7343 | 0.7111 | 0.9639 |
| 0.1153 | 3.0 | 534 | 0.1085 | {'precision': 0.7085427135678392, 'recall': 0.7768595041322314, 'f1': 0.7411300919842312, 'number': 363} | {'precision': 0.6533333333333333, 'recall': 0.7424242424242424, 'f1': 0.6950354609929078, 'number': 66} | 0.6998 | 0.7716 | 0.7339 | 0.9658 |
| 0.1153 | 4.0 | 712 | 0.1163 | {'precision': 0.6987341772151898, 'recall': 0.7603305785123967, 'f1': 0.7282321899736148, 'number': 363} | {'precision': 0.7121212121212122, 'recall': 0.7121212121212122, 'f1': 0.7121212121212122, 'number': 66} | 0.7007 | 0.7529 | 0.7258 | 0.9657 |
| 0.1153 | 5.0 | 890 | 0.1217 | {'precision': 0.7340153452685422, 'recall': 0.790633608815427, 'f1': 0.7612732095490715, 'number': 363} | {'precision': 0.6911764705882353, 'recall': 0.7121212121212122, 'f1': 0.7014925373134329, 'number': 66} | 0.7277 | 0.7786 | 0.7523 | 0.9661 |
Framework versions
- Transformers 4.38.1
- Pytorch 2.1.2+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2