metadata
license: mit
base_model: xlm-roberta-base
tags:
- generated_from_trainer
datasets:
- tweet_sentiment_multilingual
metrics:
- accuracy
- f1
model-index:
- name: scenario-TCR-4_data-cardiffnlp_tweet_sentiment_multilingual_all
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: tweet_sentiment_multilingual
type: tweet_sentiment_multilingual
config: all
split: validation
args: all
metrics:
- name: Accuracy
type: accuracy
value: 0.6257716049382716
- name: F1
type: f1
value: 0.6219727973114133
scenario-TCR-4_data-cardiffnlp_tweet_sentiment_multilingual_all
This model is a fine-tuned version of xlm-roberta-base on the tweet_sentiment_multilingual dataset. It achieves the following results on the evaluation set:
- Loss: 1.2568
- Accuracy: 0.6258
- F1: 0.6220
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: 32
- eval_batch_size: 32
- seed: 48
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 500
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 1.0005 | 1.09 | 500 | 0.9616 | 0.5517 | 0.5238 |
| 0.8578 | 2.17 | 1000 | 0.8651 | 0.5968 | 0.5750 |
| 0.7545 | 3.26 | 1500 | 0.9179 | 0.6269 | 0.6224 |
| 0.6418 | 4.35 | 2000 | 0.9266 | 0.6169 | 0.6117 |
| 0.5162 | 5.43 | 2500 | 1.0184 | 0.6331 | 0.6298 |
| 0.4188 | 6.52 | 3000 | 1.1995 | 0.6080 | 0.6075 |
| 0.3517 | 7.61 | 3500 | 1.2092 | 0.6273 | 0.6260 |
| 0.2947 | 8.7 | 4000 | 1.2915 | 0.6177 | 0.6171 |
| 0.2488 | 9.78 | 4500 | 1.3765 | 0.6219 | 0.6210 |
| 0.2152 | 10.87 | 5000 | 1.2568 | 0.6258 | 0.6220 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.1.1+cu121
- Datasets 2.14.5
- Tokenizers 0.13.3