nb-sbert-base-edu-scorer-lr3e5-bs32_swe_test

This model is a fine-tuned version of NbAiLab/nb-sbert-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.1619
• Accuracy: 0.5221
• F1 Weighted: 0.4937

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: 3e-05
• train_batch_size: 64
• eval_batch_size: 64
• 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Weighted
1.1966	1.0	1472	1.1938	0.4988	0.4553
1.1189	2.0	2944	1.2050	0.4960	0.4689
0.9824	3.0	4416	1.2919	0.4810	0.4594
0.8382	4.0	5888	1.4092	0.4726	0.4676
0.6931	5.0	7360	1.6323	0.4647	0.4557
0.5517	6.0	8832	1.8313	0.4534	0.4562
0.4569	7.0	10304	1.9833	0.4546	0.4559
0.3766	8.0	11776	2.2487	0.4397	0.4492
0.3169	9.0	13248	2.4014	0.4431	0.4492
0.2588	10.0	14720	2.5840	0.4320	0.4409
0.2246	11.0	16192	2.7939	0.4474	0.4544
0.1961	12.0	17664	2.9201	0.4540	0.4585
0.1389	13.0	19136	3.0996	0.4522	0.4590
0.1268	14.0	20608	3.2763	0.4415	0.4516
0.1016	15.0	22080	3.5122	0.4532	0.4629
0.0825	16.0	23552	3.6232	0.4603	0.4624
0.0702	17.0	25024	3.8524	0.4564	0.4636
0.0454	18.0	26496	4.0069	0.4506	0.4589
0.0414	19.0	27968	4.1465	0.4498	0.4576
0.0313	20.0	29440	4.1988	0.4556	0.4620

Framework versions

• Transformers 4.55.0
• Pytorch 2.5.1+cu121
• Datasets 4.0.0
• Tokenizers 0.21.4