roberta_test_higher_sentiment_1

This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.7852

accuracy

: 0.7919

: 0.7851

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-06
train_batch_size: 32
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 50

Training results

accuracy

| |:-------------:|:-----:|:----:|:---------------:|:------------:|:------:| | No log | 1.0 | 414 | 8.8075 | 0.2791 | 0.1218 | | 9.1738 | 2.0 | 828 | 7.7356 | 0.2814 | 0.1286 | | 7.3001 | 3.0 | 1242 | 6.6748 | 0.2814 | 0.1346 | | 6.0371 | 4.0 | 1656 | 5.6375 | 0.3012 | 0.1810 | | 4.898 | 5.0 | 2070 | 4.6651 | 0.3488 | 0.2830 | | 4.898 | 6.0 | 2484 | 3.7891 | 0.3919 | 0.3600 | | 3.8534 | 7.0 | 2898 | 3.0231 | 0.4326 | 0.4271 | | 2.9318 | 8.0 | 3312 | 2.3765 | 0.4767 | 0.4881 | | 2.2018 | 9.0 | 3726 | 1.8644 | 0.5163 | 0.5362 | | 1.6561 | 10.0 | 4140 | 1.5070 | 0.5547 | 0.5772 | | 1.3598 | 11.0 | 4554 | 1.3376 | 0.5965 | 0.6162 | | 1.3598 | 12.0 | 4968 | 1.3128 | 0.5895 | 0.6097 | | 1.2704 | 13.0 | 5382 | 1.0599 | 0.6337 | 0.6515 | | 1.1448 | 14.0 | 5796 | 0.9554 | 0.6570 | 0.6723 | | 0.9513 | 15.0 | 6210 | 0.8642 | 0.6965 | 0.7018 | | 0.8368 | 16.0 | 6624 | 0.8353 | 0.7070 | 0.7136 | | 0.7607 | 17.0 | 7038 | 0.8036 | 0.7558 | 0.7552 | | 0.7607 | 18.0 | 7452 | 0.7971 | 0.7919 | 0.7840 | | 0.6982 | 19.0 | 7866 | 0.8153 | 0.7709 | 0.7719 | | 0.6449 | 20.0 | 8280 | 0.7852 | 0.7919 | 0.7851 | | 0.5748 | 21.0 | 8694 | 0.7964 | 0.7849 | 0.7801 | | 0.545 | 22.0 | 9108 | 0.8217 | 0.7965 | 0.7861 |

Framework versions

Transformers 4.45.2
Pytorch 2.5.0+cu124
Datasets 3.0.1
Tokenizers 0.20.1

Downloads last month: -; Downloads are not tracked for this model. How to track

Safetensors

Model size

66.4M params

Tensor type

F32

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for iTroned/roberta_test_higher_sentiment_1

Base model

distilbert/distilbert-base-uncased

Finetuned

(12633)

this model