Version5ASAP_FineTuningBERT_AugV12_k3_task1_organization_k3_k3_fold2

This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7340
• Qwk: 0.5432
• Mse: 0.7329
• Rmse: 0.8561

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: 64
• eval_batch_size: 64
• 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: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	1.0	1	10.4402	0.0014	10.4404	3.2312
No log	2.0	2	10.0272	0.0	10.0273	3.1666
No log	3.0	3	9.1551	0.0	9.1552	3.0258
No log	4.0	4	8.2388	0.0	8.2390	2.8704
No log	5.0	5	7.3925	0.0	7.3928	2.7190
No log	6.0	6	6.5647	0.0	6.5651	2.5622
No log	7.0	7	5.7864	0.0589	5.7869	2.4056
No log	8.0	8	5.1557	0.0189	5.1561	2.2707
No log	9.0	9	4.6809	0.0039	4.6813	2.1636
No log	10.0	10	4.2089	0.0039	4.2094	2.0517
No log	11.0	11	3.7744	0.0	3.7748	1.9429
No log	12.0	12	3.4503	0.0	3.4508	1.8576
No log	13.0	13	3.0477	0.0	3.0482	1.7459
No log	14.0	14	2.7244	0.0	2.7248	1.6507
No log	15.0	15	2.4695	0.0336	2.4699	1.5716
No log	16.0	16	2.2525	0.1439	2.2530	1.5010
No log	17.0	17	2.0708	0.0806	2.0712	1.4392
No log	18.0	18	1.8382	0.0268	1.8386	1.3560
No log	19.0	19	1.6634	0.0107	1.6639	1.2899
No log	20.0	20	1.4985	0.0107	1.4990	1.2243
No log	21.0	21	1.3967	0.0107	1.3972	1.1820
No log	22.0	22	1.2559	0.0107	1.2563	1.1209
No log	23.0	23	1.1388	0.0107	1.1392	1.0674
No log	24.0	24	1.0484	0.0107	1.0488	1.0241
No log	25.0	25	0.9702	0.0107	0.9706	0.9852
No log	26.0	26	0.9011	0.0166	0.9015	0.9495
No log	27.0	27	0.8457	0.3033	0.8460	0.9198
No log	28.0	28	0.7920	0.3899	0.7923	0.8901
No log	29.0	29	0.7430	0.3980	0.7432	0.8621
No log	30.0	30	0.7007	0.3945	0.7009	0.8372
No log	31.0	31	0.6705	0.3742	0.6706	0.8189
No log	32.0	32	0.6416	0.3756	0.6416	0.8010
No log	33.0	33	0.6137	0.4029	0.6136	0.7834
No log	34.0	34	0.5854	0.4325	0.5852	0.7650
No log	35.0	35	0.5656	0.4424	0.5653	0.7519
No log	36.0	36	0.5724	0.4288	0.5721	0.7564
No log	37.0	37	0.5666	0.4386	0.5662	0.7525
No log	38.0	38	0.5392	0.4541	0.5387	0.7340
No log	39.0	39	0.5401	0.4423	0.5396	0.7345
No log	40.0	40	0.5443	0.4518	0.5437	0.7374
No log	41.0	41	0.5810	0.5091	0.5803	0.7618
No log	42.0	42	0.5731	0.5553	0.5724	0.7566
No log	43.0	43	0.5582	0.5870	0.5574	0.7466
No log	44.0	44	0.5697	0.5847	0.5689	0.7542
No log	45.0	45	0.6083	0.5597	0.6074	0.7794
No log	46.0	46	0.6638	0.5492	0.6628	0.8141
No log	47.0	47	0.6663	0.5506	0.6653	0.8157
No log	48.0	48	0.6912	0.5417	0.6901	0.8307
No log	49.0	49	0.6883	0.5626	0.6872	0.8290
No log	50.0	50	0.6227	0.5685	0.6217	0.7885
No log	51.0	51	0.6365	0.5678	0.6355	0.7972
No log	52.0	52	0.7207	0.5376	0.7195	0.8483
No log	53.0	53	0.7309	0.5325	0.7297	0.8542
No log	54.0	54	0.6621	0.5534	0.6610	0.8130
No log	55.0	55	0.6245	0.5655	0.6236	0.7897
No log	56.0	56	0.6443	0.5517	0.6433	0.8021
No log	57.0	57	0.7135	0.5490	0.7123	0.8440
No log	58.0	58	0.7774	0.5280	0.7761	0.8810
No log	59.0	59	0.6948	0.5446	0.6937	0.8329
No log	60.0	60	0.6093	0.5241	0.6085	0.7800
No log	61.0	61	0.6055	0.5077	0.6048	0.7777
No log	62.0	62	0.6118	0.5218	0.6109	0.7816
No log	63.0	63	0.7340	0.5432	0.7329	0.8561

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.3.1
• Tokenizers 0.21.0