metadata

language:
  - id
license: mit
base_model: indolem/indobert-base-uncased
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: sentiment-pt-pl30-4
    results: []

sentiment-pt-pl30-4

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: 0.3168
Accuracy: 0.8797
Precision: 0.8530
Recall: 0.8599
F1: 0.8563

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
0.5413	1.0	122	0.5007	0.7243	0.6593	0.6374	0.6446
0.4584	2.0	244	0.3856	0.8296	0.8123	0.7569	0.7761
0.3559	3.0	366	0.3407	0.8571	0.8638	0.7814	0.8079
0.2961	4.0	488	0.3089	0.8697	0.8438	0.8403	0.8420
0.276	5.0	610	0.2917	0.8622	0.8314	0.8425	0.8365
0.2555	6.0	732	0.2905	0.8697	0.8428	0.8428	0.8428
0.2427	7.0	854	0.3031	0.8772	0.8670	0.8281	0.8440
0.2219	8.0	976	0.2908	0.8772	0.8514	0.8531	0.8522
0.2158	9.0	1098	0.3084	0.8847	0.8760	0.8384	0.8540
0.2	10.0	1220	0.2938	0.8747	0.8457	0.8588	0.8517
0.1885	11.0	1342	0.2977	0.8772	0.8524	0.8506	0.8515
0.183	12.0	1464	0.3070	0.8847	0.8717	0.8434	0.8557
0.1752	13.0	1586	0.2959	0.8797	0.8522	0.8624	0.8570
0.1558	14.0	1708	0.3040	0.8747	0.8447	0.8638	0.8531
0.1538	15.0	1830	0.3082	0.8722	0.8431	0.8546	0.8484
0.152	16.0	1952	0.3100	0.8772	0.8576	0.8406	0.8484
0.1436	17.0	2074	0.3105	0.8747	0.8463	0.8563	0.8510
0.1426	18.0	2196	0.3119	0.8747	0.8471	0.8538	0.8503
0.1398	19.0	2318	0.3164	0.8797	0.8522	0.8624	0.8570
0.14	20.0	2440	0.3168	0.8797	0.8530	0.8599	0.8563

Framework versions

Transformers 4.39.3
Pytorch 2.3.0+cu121
Datasets 2.19.1
Tokenizers 0.15.2