File size: 2,318 Bytes
c966d38 | 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 | ---
library_name: transformers
license: mit
base_model: FacebookAI/xlm-roberta-large
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: populism_classifier_bsample_119
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. -->
# populism_classifier_bsample_119
This model is a fine-tuned version of [FacebookAI/xlm-roberta-large](https://huggingface.co/FacebookAI/xlm-roberta-large) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.8715
- Accuracy: 0.6158
- 1-f1: 0.3105
- 1-recall: 1.0
- 1-precision: 0.1838
- Balanced Acc: 0.7897
## 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: 3e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 15
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | 1-f1 | 1-recall | 1-precision | Balanced Acc |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:--------:|:-----------:|:------------:|
| 0.6817 | 1.0 | 11 | 0.8455 | 0.0865 | 0.1593 | 1.0 | 0.0865 | 0.5 |
| 0.6879 | 2.0 | 22 | 0.7567 | 0.3410 | 0.2080 | 1.0 | 0.1160 | 0.6393 |
| 0.6883 | 3.0 | 33 | 0.4163 | 0.8499 | 0.4587 | 0.7353 | 0.3333 | 0.7980 |
| 0.197 | 4.0 | 44 | 0.3296 | 0.8753 | 0.5505 | 0.8824 | 0.4 | 0.8785 |
| 0.329 | 5.0 | 55 | 0.7855 | 0.7863 | 0.44 | 0.9706 | 0.2845 | 0.8697 |
| 0.3333 | 6.0 | 66 | 1.8715 | 0.6158 | 0.3105 | 1.0 | 0.1838 | 0.7897 |
### Framework versions
- Transformers 4.46.3
- Pytorch 2.4.1+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3
|