sentiment-seq_bn-2

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: 0.3205
• Accuracy: 0.8772
• Precision: 0.8609
• Recall: 0.8356
• F1: 0.8467

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
0.5517	1.0	122	0.5131	0.7168	0.6513	0.6371	0.6424
0.4833	2.0	244	0.4657	0.7519	0.7088	0.7295	0.7159
0.4318	3.0	366	0.4056	0.8120	0.7729	0.7845	0.7781
0.3905	4.0	488	0.3811	0.8421	0.8092	0.8108	0.8100
0.3626	5.0	610	0.3652	0.8496	0.8186	0.8186	0.8186
0.3331	6.0	732	0.3646	0.8546	0.8214	0.8497	0.8325
0.3134	7.0	854	0.3440	0.8672	0.8412	0.8360	0.8385
0.2927	8.0	976	0.3412	0.8647	0.8359	0.8392	0.8376
0.2833	9.0	1098	0.3353	0.8647	0.8352	0.8417	0.8383
0.2672	10.0	1220	0.3296	0.8672	0.8367	0.8510	0.8432
0.2641	11.0	1342	0.3270	0.8772	0.8576	0.8406	0.8484
0.2549	12.0	1464	0.3352	0.8697	0.8558	0.8203	0.8350
0.2534	13.0	1586	0.3402	0.8697	0.8602	0.8153	0.8330
0.2389	14.0	1708	0.3208	0.8822	0.8574	0.8592	0.8583
0.2203	15.0	1830	0.3279	0.8747	0.8605	0.8288	0.8422
0.2298	16.0	1952	0.3175	0.8747	0.8552	0.8363	0.8448
0.2227	17.0	2074	0.3218	0.8747	0.8586	0.8313	0.8431
0.2225	18.0	2196	0.3178	0.8772	0.8524	0.8506	0.8515
0.2192	19.0	2318	0.3199	0.8772	0.8609	0.8356	0.8467
0.2229	20.0	2440	0.3205	0.8772	0.8609	0.8356	0.8467

Framework versions

• Transformers 4.40.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1