|
|
--- |
|
|
library_name: transformers |
|
|
license: mit |
|
|
base_model: deepset/gbert-large |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
metrics: |
|
|
- accuracy |
|
|
- f1 |
|
|
- precision |
|
|
- recall |
|
|
model-index: |
|
|
- name: gbert_synset_classifier_marked |
|
|
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. --> |
|
|
|
|
|
# gbert_synset_classifier_marked |
|
|
|
|
|
This model is a fine-tuned version of [deepset/gbert-large](https://huggingface.co/deepset/gbert-large) on an unknown dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.5043 |
|
|
- Accuracy: 0.8515 |
|
|
- F1: 0.8487 |
|
|
- Precision: 0.8526 |
|
|
- Recall: 0.8515 |
|
|
- F1 Macro: 0.7531 |
|
|
- Precision Macro: 0.7471 |
|
|
- Recall Macro: 0.7706 |
|
|
- F1 Micro: 0.8515 |
|
|
- Precision Micro: 0.8515 |
|
|
- Recall Micro: 0.8515 |
|
|
|
|
|
## 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: 20 |
|
|
- eval_batch_size: 20 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 4 |
|
|
- total_train_batch_size: 80 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: linear |
|
|
- lr_scheduler_warmup_steps: 50 |
|
|
- num_epochs: 5 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall | F1 Macro | Precision Macro | Recall Macro | F1 Micro | Precision Micro | Recall Micro | |
|
|
|:-------------:|:------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|:--------:|:---------------:|:------------:|:--------:|:---------------:|:------------:| |
|
|
| 2.9643 | 0.3891 | 100 | 1.2450 | 0.7537 | 0.7157 | 0.7077 | 0.7537 | 0.4231 | 0.4210 | 0.4559 | 0.7537 | 0.7537 | 0.7537 | |
|
|
| 0.9779 | 0.7782 | 200 | 0.6959 | 0.8138 | 0.8009 | 0.8022 | 0.8138 | 0.5690 | 0.5784 | 0.5873 | 0.8138 | 0.8138 | 0.8138 | |
|
|
| 0.6664 | 1.1673 | 300 | 0.5947 | 0.8313 | 0.8222 | 0.8313 | 0.8313 | 0.6503 | 0.6570 | 0.6716 | 0.8313 | 0.8313 | 0.8313 | |
|
|
| 0.5523 | 1.5564 | 400 | 0.5216 | 0.8448 | 0.8365 | 0.8405 | 0.8448 | 0.6824 | 0.6759 | 0.7057 | 0.8448 | 0.8448 | 0.8448 | |
|
|
| 0.5126 | 1.9455 | 500 | 0.5216 | 0.8484 | 0.8417 | 0.8499 | 0.8484 | 0.7004 | 0.6848 | 0.7357 | 0.8484 | 0.8484 | 0.8484 | |
|
|
| 0.4103 | 2.3346 | 600 | 0.5001 | 0.8533 | 0.8497 | 0.8579 | 0.8533 | 0.7212 | 0.7252 | 0.7375 | 0.8533 | 0.8533 | 0.8533 | |
|
|
| 0.3733 | 2.7237 | 700 | 0.5010 | 0.8466 | 0.8403 | 0.8484 | 0.8466 | 0.7325 | 0.7274 | 0.7538 | 0.8466 | 0.8466 | 0.8466 | |
|
|
| 0.3654 | 3.1128 | 800 | 0.4934 | 0.8524 | 0.8478 | 0.8547 | 0.8524 | 0.7439 | 0.7411 | 0.7630 | 0.8524 | 0.8524 | 0.8524 | |
|
|
| 0.2761 | 3.5019 | 900 | 0.5038 | 0.8533 | 0.8495 | 0.8536 | 0.8533 | 0.7611 | 0.7527 | 0.7808 | 0.8533 | 0.8533 | 0.8533 | |
|
|
| 0.274 | 3.8911 | 1000 | 0.5043 | 0.8515 | 0.8487 | 0.8526 | 0.8515 | 0.7531 | 0.7471 | 0.7706 | 0.8515 | 0.8515 | 0.8515 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.45.2 |
|
|
- Pytorch 2.3.1+cu121 |
|
|
- Datasets 2.20.0 |
|
|
- Tokenizers 0.20.3 |
|
|
|
