File size: 2,007 Bytes
336e735 |
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 |
---
license: cc-by-4.0
base_model: EMBEDDIA/crosloengual-bert
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: fine_tuned_copa_croslo
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. -->
# fine_tuned_copa_croslo
This model is a fine-tuned version of [EMBEDDIA/crosloengual-bert](https://huggingface.co/EMBEDDIA/crosloengual-bert) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6931
- Accuracy: 0.55
- F1: 0.3903
## 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: 0.003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 400
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
| 0.7205 | 1.0 | 50 | 0.6931 | 0.55 | 0.3903 |
| 0.7204 | 2.0 | 100 | 0.6931 | 0.55 | 0.3903 |
| 0.7044 | 3.0 | 150 | 0.6931 | 0.55 | 0.3903 |
| 0.7288 | 4.0 | 200 | 0.6931 | 0.55 | 0.3903 |
| 0.7268 | 5.0 | 250 | 0.6931 | 0.54 | 0.3857 |
| 0.7047 | 6.0 | 300 | 0.6931 | 0.55 | 0.3903 |
| 0.7291 | 7.0 | 350 | 0.6931 | 0.55 | 0.3903 |
| 0.713 | 8.0 | 400 | 0.6931 | 0.55 | 0.3903 |
### Framework versions
- Transformers 4.40.1
- Pytorch 2.1.1+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1
|