---
library_name: transformers
license: mit
base_model: FacebookAI/roberta-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: sentiment-analysis-roberta-base-V1.5_ima_ds
  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. -->

# sentiment-analysis-roberta-base-V1.5_ima_ds

This model is a fine-tuned version of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.3346
- Accuracy: 0.5869
- Precision Macro: 0.5474
- Recall Macro: 0.5539
- F1 Macro: 0.5472
- Precision Weighted: 0.5941
- Recall Weighted: 0.5869
- F1 Weighted: 0.5860

## 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: 8
- seed: 3407
- optimizer: Use OptimizerNames.ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 50
- num_epochs: 20

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision Macro | Recall Macro | F1 Macro | Precision Weighted | Recall Weighted | F1 Weighted |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------------:|:------------:|:--------:|:------------------:|:---------------:|:-----------:|
| 1.5983        | 0.4   | 20   | 1.6093          | 0.2418   | 0.0484          | 0.2          | 0.0779   | 0.0585             | 0.2418          | 0.0942      |
| 1.6121        | 0.8   | 40   | 1.6067          | 0.2418   | 0.0484          | 0.2          | 0.0779   | 0.0585             | 0.2418          | 0.0942      |
| 1.5924        | 1.2   | 60   | 1.5340          | 0.4358   | 0.3628          | 0.3576       | 0.3453   | 0.4328             | 0.4358          | 0.4141      |
| 1.3138        | 1.6   | 80   | 1.3488          | 0.4736   | 0.4587          | 0.4645       | 0.4236   | 0.5236             | 0.4736          | 0.4577      |
| 1.2226        | 2.0   | 100  | 1.2556          | 0.4786   | 0.4708          | 0.4714       | 0.4551   | 0.5293             | 0.4786          | 0.4958      |
| 0.7537        | 2.4   | 120  | 1.1485          | 0.5718   | 0.5461          | 0.5623       | 0.5453   | 0.5897             | 0.5718          | 0.5754      |
| 0.9658        | 2.8   | 140  | 1.1688          | 0.5466   | 0.5373          | 0.5496       | 0.5230   | 0.6096             | 0.5466          | 0.5643      |
| 0.7872        | 3.2   | 160  | 1.1887          | 0.5718   | 0.5398          | 0.5341       | 0.5285   | 0.5901             | 0.5718          | 0.5756      |
| 1.1199        | 3.6   | 180  | 1.1970          | 0.5743   | 0.5774          | 0.5609       | 0.5404   | 0.6265             | 0.5743          | 0.5847      |
| 1.0137        | 4.0   | 200  | 1.2344          | 0.4987   | 0.5382          | 0.5293       | 0.4768   | 0.6059             | 0.4987          | 0.5158      |
| 0.6547        | 4.4   | 220  | 1.1846          | 0.5919   | 0.5737          | 0.5997       | 0.5751   | 0.6139             | 0.5919          | 0.5972      |
| 0.7685        | 4.8   | 240  | 1.2460          | 0.5970   | 0.5759          | 0.6021       | 0.5749   | 0.6252             | 0.5970          | 0.6035      |
| 0.5167        | 5.2   | 260  | 1.2670          | 0.5919   | 0.5863          | 0.5959       | 0.5697   | 0.6268             | 0.5919          | 0.5980      |
| 0.6995        | 5.6   | 280  | 1.3015          | 0.5743   | 0.5686          | 0.5830       | 0.5497   | 0.6167             | 0.5743          | 0.5775      |
| 0.8019        | 6.0   | 300  | 1.3346          | 0.5869   | 0.5474          | 0.5539       | 0.5472   | 0.5941             | 0.5869          | 0.5860      |


### Framework versions

- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.1