IndoBERT-Sentiment-Analysis7

This model is a fine-tuned version of indobenchmark/indobert-base-p1 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3947
• Accuracy: 0.9359
• F1 Score: 0.9357

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: 1e-05
• train_batch_size: 6
• eval_batch_size: 6
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Score
0.6214	0.1096	50	0.6184	0.6513	0.6503
0.5834	0.2193	100	0.5604	0.6962	0.6923
0.5366	0.3289	150	0.4462	0.8090	0.8089
0.4827	0.4386	200	0.4422	0.8141	0.8131
0.3999	0.5482	250	0.4463	0.8295	0.8295
0.4793	0.6579	300	0.3855	0.8551	0.8550
0.3465	0.7675	350	0.4340	0.8513	0.8506
0.329	0.8772	400	0.4847	0.8615	0.8612
0.3782	0.9868	450	0.6790	0.7949	0.7884
0.286	1.0965	500	0.4429	0.8897	0.8897
0.276	1.2061	550	0.4755	0.8974	0.8972
0.3092	1.3158	600	0.6285	0.8551	0.8532
0.1986	1.4254	650	0.4787	0.8974	0.8970
0.2838	1.5351	700	0.3770	0.9205	0.9204
0.3431	1.6447	750	0.4089	0.9051	0.9047
0.1427	1.7544	800	0.3640	0.9179	0.9178
0.1827	1.8640	850	0.5831	0.8808	0.8796
0.1874	1.9737	900	0.4004	0.9205	0.9202
0.1265	2.0833	950	0.4419	0.9128	0.9124
0.1044	2.1930	1000	0.3590	0.9269	0.9268
0.0647	2.3026	1050	0.3908	0.9282	0.9280
0.0609	2.4123	1100	0.3832	0.9321	0.9319
0.049	2.5219	1150	0.5279	0.9064	0.9059
0.1219	2.6316	1200	0.4060	0.9346	0.9345
0.1665	2.7412	1250	0.3126	0.9359	0.9358
0.1013	2.8509	1300	0.2925	0.9487	0.9487
0.1665	2.9605	1350	0.3980	0.9269	0.9267
0.1647	3.0702	1400	0.3481	0.9346	0.9344
0.0637	3.1798	1450	0.4226	0.9256	0.9253
0.0563	3.2895	1500	0.4031	0.9308	0.9306
0.031	3.3991	1550	0.3697	0.9385	0.9383
0.0254	3.5088	1600	0.3933	0.9359	0.9357
0.0792	3.6184	1650	0.3147	0.9474	0.9474
0.0364	3.7281	1700	0.4430	0.9269	0.9267
0.0672	3.8377	1750	0.3703	0.9372	0.9370
0.0633	3.9474	1800	0.4756	0.9192	0.9189
0.046	4.0570	1850	0.3599	0.9449	0.9448
0.0758	4.1667	1900	0.4557	0.9256	0.9254
0.0219	4.2763	1950	0.4249	0.9269	0.9267
0.0043	4.3860	2000	0.4690	0.9244	0.9240
0.0651	4.4956	2050	0.4021	0.9333	0.9331
0.0854	4.6053	2100	0.3757	0.9385	0.9383
0.0377	4.7149	2150	0.4022	0.9333	0.9331
0.0066	4.8246	2200	0.3904	0.9372	0.9370
0.0021	4.9342	2250	0.3943	0.9359	0.9357

Framework versions

• Transformers 4.53.2
• Pytorch 2.6.0+cu124
• Datasets 2.14.4
• Tokenizers 0.21.2