|
--- |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- enoriega/odinsynth_dataset |
|
model-index: |
|
- name: rule_learning_margin_1mm_spanpred |
|
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. --> |
|
|
|
# rule_learning_margin_1mm_spanpred |
|
|
|
This model is a fine-tuned version of [enoriega/rule_softmatching](https://huggingface.co/enoriega/rule_softmatching) on the enoriega/odinsynth_dataset dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3250 |
|
- Margin Accuracy: 0.8518 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 4 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 2000 |
|
- total_train_batch_size: 8000 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 3.0 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Margin Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------------:| |
|
| 0.5448 | 0.16 | 20 | 0.5229 | 0.7717 | |
|
| 0.4571 | 0.32 | 40 | 0.4292 | 0.8109 | |
|
| 0.4296 | 0.48 | 60 | 0.4009 | 0.8193 | |
|
| 0.4028 | 0.64 | 80 | 0.3855 | 0.8296 | |
|
| 0.3878 | 0.8 | 100 | 0.3757 | 0.8334 | |
|
| 0.3831 | 0.96 | 120 | 0.3643 | 0.8367 | |
|
| 0.3591 | 1.12 | 140 | 0.3582 | 0.8393 | |
|
| 0.3598 | 1.28 | 160 | 0.3533 | 0.8401 | |
|
| 0.3635 | 1.44 | 180 | 0.3442 | 0.8427 | |
|
| 0.3478 | 1.6 | 200 | 0.3406 | 0.8472 | |
|
| 0.342 | 1.76 | 220 | 0.3352 | 0.8479 | |
|
| 0.3327 | 1.92 | 240 | 0.3352 | 0.8486 | |
|
| 0.3487 | 2.08 | 260 | 0.3293 | 0.8487 | |
|
| 0.3387 | 2.24 | 280 | 0.3298 | 0.8496 | |
|
| 0.3457 | 2.4 | 300 | 0.3279 | 0.8505 | |
|
| 0.3483 | 2.56 | 320 | 0.3286 | 0.8510 | |
|
| 0.3421 | 2.72 | 340 | 0.3245 | 0.8517 | |
|
| 0.3332 | 2.88 | 360 | 0.3252 | 0.8517 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.19.2 |
|
- Pytorch 1.11.0 |
|
- Datasets 2.2.1 |
|
- Tokenizers 0.12.1 |
|
|
