RoBERTa-Base-SE2025T11A-sun-v20250107143556

This model is a fine-tuned version of w11wo/sundanese-roberta-base-emotion-classifier on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3533
• F1 Macro: 0.6178
• F1 Micro: 0.6190
• F1 Weighted: 0.6165
• F1 Samples: 0.6113
• F1 Label Marah: 0.6408
• F1 Label Jijik: 0.5435
• F1 Label Takut: 0.5938
• F1 Label Senang: 0.8395
• F1 Label Sedih: 0.5758
• F1 Label Terkejut: 0.5063
• F1 Label Biasa: 0.625

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: 2e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Use 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	F1 Macro	F1 Micro	F1 Weighted	F1 Samples	F1 Label Marah	F1 Label Jijik	F1 Label Takut	F1 Label Senang	F1 Label Sedih	F1 Label Terkejut	F1 Label Biasa
0.502	0.1805	100	0.4277	0.1484	0.2553	0.1776	0.1892	0.5255	0.04	0.0	0.2917	0.1818	0.0	0.0
0.4765	0.3610	200	0.4058	0.2881	0.3526	0.3134	0.2492	0.5652	0.0755	0.48	0.4727	0.4231	0.0	0.0
0.4129	0.5415	300	0.3793	0.2532	0.3228	0.2553	0.2177	0.1356	0.0	0.4348	0.7949	0.4074	0.0	0.0
0.4055	0.7220	400	0.3522	0.3939	0.4559	0.4206	0.3194	0.4675	0.2069	0.5	0.7765	0.4727	0.3333	0.0
0.4033	0.9025	500	0.3440	0.3094	0.4011	0.3305	0.2988	0.5116	0.0	0.4681	0.7711	0.4151	0.0	0.0
0.342	1.0830	600	0.3285	0.4847	0.5581	0.5235	0.4581	0.6263	0.2985	0.52	0.8140	0.5758	0.5581	0.0
0.3255	1.2635	700	0.3483	0.4836	0.5529	0.5265	0.4913	0.6604	0.4110	0.5753	0.7901	0.4528	0.4952	0.0
0.285	1.4440	800	0.3312	0.5159	0.5356	0.5193	0.4790	0.4324	0.4773	0.5769	0.8046	0.5352	0.4	0.3846
0.323	1.6245	900	0.3123	0.5834	0.5833	0.5651	0.5461	0.56	0.3235	0.6452	0.8293	0.6053	0.4333	0.6875
0.3056	1.8051	1000	0.3225	0.6087	0.6	0.5863	0.5420	0.5122	0.3478	0.6471	0.8395	0.6269	0.56	0.7273
0.2933	1.9856	1100	0.3183	0.5223	0.5691	0.5524	0.5315	0.6733	0.4048	0.5490	0.7727	0.5625	0.5122	0.1818
0.2083	2.1661	1200	0.3204	0.5837	0.5904	0.5868	0.5518	0.5647	0.5208	0.6316	0.8205	0.4912	0.5567	0.5
0.225	2.3466	1300	0.3042	0.6107	0.6201	0.6068	0.5941	0.6465	0.4595	0.7222	0.8293	0.5397	0.4776	0.6
0.21	2.5271	1400	0.3257	0.6022	0.6008	0.5980	0.5884	0.62	0.5370	0.5818	0.7901	0.5484	0.4928	0.6452
0.1873	2.7076	1500	0.3303	0.6000	0.6100	0.6033	0.6102	0.6538	0.5495	0.5862	0.8043	0.5714	0.4634	0.5714
0.2102	2.8881	1600	0.3205	0.5906	0.6055	0.5977	0.6005	0.6783	0.4944	0.5862	0.7857	0.5714	0.5	0.5185
0.186	3.0686	1700	0.3239	0.6094	0.6160	0.6141	0.5887	0.6237	0.5714	0.5758	0.8354	0.5484	0.5591	0.5517
0.142	3.2491	1800	0.3281	0.6074	0.6154	0.6108	0.5992	0.6486	0.5060	0.6061	0.8395	0.5714	0.5287	0.5517
0.1476	3.4296	1900	0.3395	0.6229	0.6157	0.6123	0.5971	0.625	0.5	0.5660	0.8608	0.6061	0.48	0.7222
0.1473	3.6101	2000	0.3463	0.6006	0.6163	0.6118	0.6059	0.625	0.5455	0.5574	0.8395	0.5938	0.5814	0.4615
0.1239	3.7906	2100	0.3511	0.6009	0.6024	0.5975	0.5944	0.6465	0.5055	0.5517	0.7907	0.5714	0.4935	0.6471
0.1392	3.9711	2200	0.3458	0.6326	0.6340	0.6313	0.6258	0.6602	0.5769	0.5424	0.8193	0.5846	0.5783	0.6667
0.1247	4.1516	2300	0.3484	0.6218	0.6216	0.6188	0.6129	0.6139	0.5625	0.5614	0.8293	0.5846	0.5542	0.6471
0.0873	4.3321	2400	0.3492	0.6195	0.6248	0.6229	0.6198	0.6465	0.5657	0.5846	0.8293	0.5758	0.5542	0.5806
0.0979	4.5126	2500	0.3490	0.6215	0.6211	0.6197	0.6072	0.6275	0.5517	0.5970	0.8462	0.5846	0.5185	0.625
0.1132	4.6931	2600	0.3534	0.6235	0.6236	0.6220	0.6162	0.6408	0.5376	0.6061	0.8293	0.5846	0.5412	0.625
0.0915	4.8736	2700	0.3533	0.6178	0.6190	0.6165	0.6113	0.6408	0.5435	0.5938	0.8395	0.5758	0.5063	0.625

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0