results

This model is a fine-tuned version of distilbert-base-cased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.3395
• Accuracy: 0.6088
• F1 Weighted: 0.6068

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: 10
• eval_batch_size: 16
• 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
• lr_scheduler_warmup_steps: 100
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Weighted
1.9969	0.1562	100	1.7350	0.4181	0.3685
1.5932	0.3125	200	1.3946	0.5038	0.4816
1.3938	0.4688	300	1.3229	0.5312	0.5406
1.3342	0.625	400	1.2797	0.5369	0.5280
1.2588	0.7812	500	1.2016	0.5737	0.5678
1.3359	0.9375	600	1.2025	0.5737	0.5472
1.0521	1.0938	700	1.2580	0.5531	0.5427
0.947	1.25	800	1.2156	0.5613	0.5656
1.0765	1.4062	900	1.1659	0.6044	0.6011
0.9336	1.5625	1000	1.2022	0.5781	0.5773
1.0288	1.7188	1100	1.1642	0.5756	0.5877
0.9665	1.875	1200	1.1635	0.5781	0.5821
0.8987	2.0312	1300	1.1913	0.5969	0.5982
0.6252	2.1875	1400	1.2783	0.595	0.5948
0.5977	2.3438	1500	1.2460	0.5938	0.5894
0.5646	2.5	1600	1.3038	0.5844	0.5915
0.6488	2.6562	1700	1.2850	0.5925	0.5955
0.627	2.8125	1800	1.2690	0.59	0.5927
0.6441	2.9688	1900	1.3395	0.6088	0.6068
0.3832	3.125	2000	1.4401	0.6088	0.6092
0.3338	3.2812	2100	1.5685	0.5831	0.5864
0.3475	3.4375	2200	1.6456	0.5806	0.5846
0.4362	3.5938	2300	1.5581	0.5825	0.5890
0.3565	3.75	2400	1.6010	0.5981	0.5993
0.3958	3.9062	2500	1.6087	0.5938	0.5944
0.2844	4.0625	2600	1.6917	0.5994	0.5980
0.174	4.2188	2700	1.8947	0.5906	0.5956
0.2393	4.375	2800	1.9103	0.5894	0.5897
0.2019	4.5312	2900	2.0275	0.5819	0.5854
0.1895	4.6875	3000	1.9962	0.5962	0.5935
0.2885	4.8438	3100	2.0387	0.5944	0.5932
0.2672	5.0	3200	2.0070	0.595	0.5938
0.1089	5.1562	3300	2.2210	0.5919	0.5945
0.1114	5.3125	3400	2.3073	0.5863	0.5884
0.1274	5.4688	3500	2.3061	0.5994	0.5994
0.1403	5.625	3600	2.2753	0.5894	0.5932
0.1869	5.7812	3700	2.2661	0.5925	0.5935
0.1769	5.9375	3800	2.2007	0.5975	0.6016
0.129	6.0938	3900	2.2289	0.6075	0.6100
0.0945	6.25	4000	2.3460	0.6038	0.6080
0.0913	6.4062	4100	2.4089	0.6038	0.6060
0.111	6.5625	4200	2.3776	0.6012	0.6039
0.1355	6.7188	4300	2.3579	0.6069	0.6069
0.1182	6.875	4400	2.3727	0.6012	0.6050
0.1049	7.0312	4500	2.4246	0.6069	0.6100
0.0802	7.1875	4600	2.5167	0.5988	0.6046
0.0665	7.3438	4700	2.5161	0.605	0.6060
0.0906	7.5	4800	2.5229	0.6088	0.6166
0.0781	7.6562	4900	2.5169	0.5994	0.5970
0.0689	7.8125	5000	2.5068	0.6	0.5987
0.1288	7.9688	5100	2.5147	0.5925	0.5974
0.0602	8.125	5200	2.5465	0.6	0.6045
0.0507	8.2812	5300	2.5416	0.605	0.6079
0.0589	8.4375	5400	2.5926	0.5962	0.6013
0.0446	8.5938	5500	2.5855	0.6062	0.6079
0.0994	8.75	5600	2.5714	0.6056	0.6097
0.0883	8.9062	5700	2.5625	0.6088	0.6123
0.0495	9.0625	5800	2.5795	0.6062	0.6095
0.0321	9.2188	5900	2.5991	0.6006	0.6045
0.0498	9.375	6000	2.5928	0.6038	0.6062
0.0303	9.5312	6100	2.5942	0.6056	0.6085
0.0552	9.6875	6200	2.5930	0.6069	0.6099
0.0394	9.8438	6300	2.5990	0.605	0.6076
0.0645	10.0	6400	2.5997	0.6056	0.6088

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu124
• Datasets 4.5.0
• Tokenizers 0.21.0