sentiment-seq_bn-rf64-0

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: 0.3469
• Accuracy: 0.8471
• Precision: 0.8147
• Recall: 0.8193
• F1: 0.8169

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
0.56	1.0	122	0.5053	0.6992	0.6181	0.5922	0.5970
0.5012	2.0	244	0.4995	0.7393	0.7072	0.7406	0.7129
0.4698	3.0	366	0.4474	0.7794	0.7363	0.7515	0.7426
0.43	4.0	488	0.4073	0.8145	0.7774	0.7688	0.7728
0.4124	5.0	610	0.4316	0.7820	0.7467	0.7807	0.7561
0.3812	6.0	732	0.4159	0.7920	0.7556	0.7878	0.7655
0.3634	7.0	854	0.3742	0.8371	0.8032	0.8047	0.8040
0.3458	8.0	976	0.3700	0.8371	0.8028	0.8072	0.8049
0.3346	9.0	1098	0.3915	0.8170	0.7796	0.8031	0.7888
0.3173	10.0	1220	0.3670	0.8346	0.8005	0.8005	0.8005
0.3133	11.0	1342	0.3597	0.8471	0.8167	0.8118	0.8142
0.3174	12.0	1464	0.3640	0.8371	0.8021	0.8122	0.8068
0.3056	13.0	1586	0.3489	0.8546	0.8284	0.8146	0.8210
0.3018	14.0	1708	0.3514	0.8521	0.8237	0.8154	0.8193
0.2979	15.0	1830	0.3517	0.8396	0.8060	0.8090	0.8075
0.2818	16.0	1952	0.3599	0.8321	0.7961	0.8087	0.8018
0.2918	17.0	2074	0.3496	0.8471	0.8153	0.8168	0.8160
0.2921	18.0	2196	0.3558	0.8371	0.8019	0.8147	0.8077
0.2807	19.0	2318	0.3456	0.8496	0.8180	0.8211	0.8195
0.2793	20.0	2440	0.3469	0.8471	0.8147	0.8193	0.8169

Framework versions

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