metadata

license: mit
base_model: indolem/indobert-base-uncased
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: sentiment-pt-pl10-4
    results: []

sentiment-pt-pl10-4

This model is a fine-tuned version of indolem/indobert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.3026
Accuracy: 0.8822
Precision: 0.8624
Recall: 0.8492
F1: 0.8553

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: 30
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 20.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.5514	1.0	122	0.5084	0.7218	0.6601	0.6507	0.6546
0.4753	2.0	244	0.4007	0.8170	0.7870	0.7530	0.7662
0.3834	3.0	366	0.3542	0.8396	0.8449	0.7540	0.7805
0.3188	4.0	488	0.3214	0.8622	0.8342	0.8325	0.8333
0.2981	5.0	610	0.2984	0.8822	0.8624	0.8492	0.8553
0.2835	6.0	732	0.2810	0.8697	0.8520	0.8253	0.8368
0.2517	7.0	854	0.2866	0.8872	0.8672	0.8577	0.8622
0.2374	8.0	976	0.2997	0.8797	0.8671	0.8349	0.8485
0.2293	9.0	1098	0.2909	0.8797	0.8600	0.8449	0.8518
0.2091	10.0	1220	0.2928	0.8822	0.8564	0.8617	0.8590
0.198	11.0	1342	0.2847	0.8797	0.8522	0.8624	0.8570
0.1906	12.0	1464	0.3120	0.8747	0.8586	0.8313	0.8431
0.1818	13.0	1586	0.2906	0.8772	0.8535	0.8481	0.8507
0.1756	14.0	1708	0.2810	0.8772	0.8524	0.8506	0.8515
0.174	15.0	1830	0.2829	0.8847	0.8634	0.8559	0.8595
0.1705	16.0	1952	0.2922	0.8822	0.8624	0.8492	0.8553
0.1509	17.0	2074	0.2991	0.8822	0.8596	0.8542	0.8568
0.1549	18.0	2196	0.3000	0.8822	0.8624	0.8492	0.8553
0.1469	19.0	2318	0.2943	0.8847	0.8609	0.8609	0.8609
0.1493	20.0	2440	0.3026	0.8822	0.8624	0.8492	0.8553

Framework versions

Transformers 4.39.3
Pytorch 2.3.0+cu121
Datasets 2.19.1
Tokenizers 0.15.2