metadata
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: rule_learning_margin_1mm
results: []
rule_learning_margin_1mm
This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3813
- Margin Accuracy: 0.8240
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 |
|---|---|---|---|---|
| 0.6482 | 0.16 | 20 | 0.6494 | 0.7263 |
| 0.5151 | 0.32 | 40 | 0.5088 | 0.7792 |
| 0.4822 | 0.48 | 60 | 0.4429 | 0.8045 |
| 0.4472 | 0.64 | 80 | 0.4265 | 0.8107 |
| 0.4352 | 0.8 | 100 | 0.4155 | 0.8132 |
| 0.4335 | 0.96 | 120 | 0.4128 | 0.8116 |
| 0.4113 | 1.12 | 140 | 0.4119 | 0.8142 |
| 0.4186 | 1.28 | 160 | 0.4075 | 0.8120 |
| 0.42 | 1.44 | 180 | 0.4072 | 0.8123 |
| 0.4175 | 1.6 | 200 | 0.4080 | 0.8130 |
| 0.4097 | 1.76 | 220 | 0.4031 | 0.8128 |
| 0.397 | 1.92 | 240 | 0.4004 | 0.8130 |
| 0.4115 | 2.08 | 260 | 0.3979 | 0.8136 |
| 0.4108 | 2.24 | 280 | 0.3940 | 0.8167 |
| 0.4125 | 2.4 | 300 | 0.3879 | 0.8218 |
| 0.4117 | 2.56 | 320 | 0.3848 | 0.8217 |
| 0.3967 | 2.72 | 340 | 0.3818 | 0.8231 |
| 0.3947 | 2.88 | 360 | 0.3813 | 0.8240 |
Framework versions
- Transformers 4.19.2
- Pytorch 1.11.0
- Datasets 2.2.1
- Tokenizers 0.12.1