Downloads · 30 days
6
16% of all-time downloads
gerskill-gbert-job-extended
dathi103/gerskill-gbert-job-extended
gerskill-gbert-job-extended is a token classification model from dathi103. 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
37
Public
Parameters
112M
6.3 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors448 MB · 99%
At a glance
- Task
- Token Classification
- Library
- transformers
- License
- mit
- Model type
- bert
- Access
- Public
- Created
- Jun 7, 2024
- Updated
- Jun 8, 2024
- SHA
- 2a6de260
Try a prompt
Base models
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. -->
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: 0.0806
- Hard: {'precision': 0.7995867768595041, 'recall': 0.8524229074889867, 'f1': 0.8251599147121534, 'number': 454}
- Soft: {'precision': 0.7804878048780488, 'recall': 0.7804878048780488, 'f1': 0.7804878048780488, 'number': 82}
- Overall Precision: 0.7968
- Overall Recall: 0.8414
- Overall F1: 0.8185
- Overall Accuracy: 0.9750
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': 0.6715867158671587, 'recall': 0.801762114537445, 'f1': 0.7309236947791165, 'number': 454} | {'precision': 0.6105263157894737, 'recall': 0.7073170731707317, 'f1': 0.6553672316384181, 'number': 82} | 0.6625 | 0.7873 | 0.7195 | 0.9592 |
| No log | 2.0 | 356 | 0.0762 | {'precision': 0.7698744769874477, 'recall': 0.8105726872246696, 'f1': 0.7896995708154506, 'number': 454} | {'precision': 0.7532467532467533, 'recall': 0.7073170731707317, 'f1': 0.7295597484276729, 'number': 82} | 0.7676 | 0.7948 | 0.7809 | 0.9705 |
| 0.1183 | 3.0 | 534 | 0.0713 | {'precision': 0.7958762886597938, 'recall': 0.8502202643171806, 'f1': 0.8221512247071352, 'number': 454} | {'precision': 0.7974683544303798, 'recall': 0.7682926829268293, 'f1': 0.782608695652174, 'number': 82} | 0.7961 | 0.8377 | 0.8164 | 0.9735 |
| 0.1183 | 4.0 | 712 | 0.0785 | {'precision': 0.7962962962962963, 'recall': 0.8524229074889867, 'f1': 0.823404255319149, 'number': 454} | {'precision': 0.7901234567901234, 'recall': 0.7804878048780488, 'f1': 0.7852760736196319, 'number': 82} | 0.7954 | 0.8414 | 0.8178 | 0.9739 |
| 0.1183 | 5.0 | 890 | 0.0806 | {'precision': 0.7995867768595041, 'recall': 0.8524229074889867, 'f1': 0.8251599147121534, 'number': 454} | {'precision': 0.7804878048780488, 'recall': 0.7804878048780488, 'f1': 0.7804878048780488, 'number': 82} | 0.7968 | 0.8414 | 0.8185 | 0.9750 |
Framework versions
- Transformers 4.38.1
- Pytorch 2.1.2+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Task
- Token Classification
- Library
- transformers
- Type
- bert
- License
- mit
- Created
- Jun 7, 2024
- Updated
- Jun 8, 2024