File size: 2,213 Bytes
0dbb101 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 |
---
base_model: yihongLiu/furina
tags:
- generated_from_trainer
model-index:
- name: furina_seed42_eng_amh_esp_cross_2e-05
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. -->
# furina_seed42_eng_amh_esp_cross_2e-05
This model is a fine-tuned version of [yihongLiu/furina](https://huggingface.co/yihongLiu/furina) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0188
- Spearman Corr: 0.7538
## 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: 128
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 30
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Spearman Corr |
|:-------------:|:-----:|:----:|:---------------:|:-------------:|
| No log | 0.53 | 200 | 0.0279 | 0.6644 |
| No log | 1.06 | 400 | 0.0165 | 0.7300 |
| No log | 1.59 | 600 | 0.0170 | 0.7252 |
| 0.0434 | 2.12 | 800 | 0.0157 | 0.7411 |
| 0.0434 | 2.65 | 1000 | 0.0160 | 0.7515 |
| 0.0434 | 3.17 | 1200 | 0.0162 | 0.7554 |
| 0.0434 | 3.7 | 1400 | 0.0160 | 0.7602 |
| 0.019 | 4.23 | 1600 | 0.0170 | 0.7490 |
| 0.019 | 4.76 | 1800 | 0.0169 | 0.7468 |
| 0.019 | 5.29 | 2000 | 0.0165 | 0.7473 |
| 0.019 | 5.82 | 2200 | 0.0181 | 0.7492 |
| 0.0138 | 6.35 | 2400 | 0.0188 | 0.7538 |
### Framework versions
- Transformers 4.37.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2
|