IndoBERT-Sentiment-Analysis

This model is a fine-tuned version of indobenchmark/indobert-base-p1 on the indonlu dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4221
  • Accuracy: 0.9452
  • F1 Score: 0.9451

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: 6
  • eval_batch_size: 6
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 5

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Score
0.3499 0.27 500 0.2392 0.9310 0.9311
0.3181 0.55 1000 0.3354 0.9175 0.9158
0.3001 0.82 1500 0.2965 0.9238 0.9243
0.2534 1.09 2000 0.3513 0.9222 0.9218
0.1692 1.36 2500 0.2657 0.9405 0.9399
0.1543 1.64 3000 0.4046 0.9198 0.9191
0.1827 1.91 3500 0.2800 0.9317 0.9319
0.1061 2.18 4000 0.3352 0.9389 0.9389
0.0639 2.45 4500 0.4033 0.9373 0.9365
0.0709 2.73 5000 0.3508 0.9365 0.9360
0.0922 3.0 5500 0.3313 0.9397 0.9394
0.0274 3.27 6000 0.3635 0.9444 0.9440
0.0273 3.54 6500 0.4074 0.9389 0.9387
0.0414 3.82 7000 0.3863 0.9405 0.9405
0.0156 4.09 7500 0.4128 0.9413 0.9412
0.0067 4.36 8000 0.4469 0.9397 0.9399
0.0056 4.63 8500 0.4297 0.9444 0.9445
0.0124 4.91 9000 0.4227 0.9452 0.9451

Framework versions

  • Transformers 4.39.0.dev0
  • Pytorch 2.1.0.dev20230729
  • Datasets 2.14.0
  • Tokenizers 0.15.2
Downloads last month
533
Safetensors
Model size
124M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for crypter70/IndoBERT-Sentiment-Analysis

Finetuned
(29)
this model

Dataset used to train crypter70/IndoBERT-Sentiment-Analysis

Space using crypter70/IndoBERT-Sentiment-Analysis 1

Evaluation results