Version3ASAP_FineTuningBERT_AugV12_k5_task1_organization_k5_k5_fold1

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: 0.7538
• Qwk: 0.6266
• Mse: 0.7534
• Rmse: 0.8680

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	2	9.7241	0.0	9.7217	3.1180
No log	2.0	4	9.2016	0.0	9.1992	3.0330
No log	3.0	6	8.3283	0.0	8.3260	2.8855
No log	4.0	8	6.9089	0.0	6.9066	2.6280
No log	5.0	10	5.3850	0.0322	5.3829	2.3201
No log	6.0	12	4.1929	0.0079	4.1908	2.0472
No log	7.0	14	3.3207	0.0	3.3187	1.8217
No log	8.0	16	2.8357	0.0	2.8339	1.6834
No log	9.0	18	2.2620	0.1721	2.2601	1.5034
No log	10.0	20	1.7518	0.0482	1.7501	1.3229
No log	11.0	22	1.4853	0.0	1.4838	1.2181
No log	12.0	24	1.1288	0.0	1.1274	1.0618
No log	13.0	26	0.9611	0.0	0.9597	0.9797
No log	14.0	28	0.9226	0.1287	0.9214	0.9599
No log	15.0	30	0.8485	0.0784	0.8473	0.9205
No log	16.0	32	0.8796	0.0660	0.8786	0.9373
No log	17.0	34	0.9571	0.0660	0.9562	0.9778
No log	18.0	36	0.8815	0.1155	0.8806	0.9384
No log	19.0	38	1.0250	0.3560	1.0243	1.0121
No log	20.0	40	0.8997	0.4176	0.8989	0.9481
No log	21.0	42	0.7637	0.4738	0.7630	0.8735
No log	22.0	44	1.0284	0.4218	1.0279	1.0138
No log	23.0	46	0.5614	0.6122	0.5607	0.7488
No log	24.0	48	0.4625	0.6427	0.4619	0.6796
No log	25.0	50	0.8144	0.5907	0.8140	0.9022
No log	26.0	52	0.4296	0.7126	0.4290	0.6550
No log	27.0	54	0.5208	0.6802	0.5204	0.7214
No log	28.0	56	1.2385	0.5301	1.2383	1.1128
No log	29.0	58	0.5258	0.6850	0.5254	0.7248
No log	30.0	60	0.5192	0.7047	0.5187	0.7202
No log	31.0	62	0.9142	0.6134	0.9140	0.9560
No log	32.0	64	0.4861	0.7137	0.4856	0.6969
No log	33.0	66	0.6462	0.6628	0.6457	0.8036
No log	34.0	68	1.4212	0.5138	1.4209	1.1920
No log	35.0	70	0.7236	0.6381	0.7231	0.8504
No log	36.0	72	0.7302	0.6425	0.7297	0.8543
No log	37.0	74	1.1707	0.5470	1.1704	1.0819
No log	38.0	76	0.5452	0.6978	0.5447	0.7381
No log	39.0	78	0.5341	0.7089	0.5337	0.7305
No log	40.0	80	1.4149	0.5133	1.4148	1.1894
No log	41.0	82	1.7980	0.4546	1.7979	1.3408
No log	42.0	84	0.8995	0.6082	0.8992	0.9482
No log	43.0	86	0.7184	0.6500	0.7180	0.8474
No log	44.0	88	1.1911	0.5454	1.1909	1.0913
No log	45.0	90	1.1181	0.5487	1.1178	1.0573
No log	46.0	92	0.5903	0.7003	0.5899	0.7680
No log	47.0	94	0.5930	0.6970	0.5926	0.7698
No log	48.0	96	1.1001	0.5489	1.0998	1.0487
No log	49.0	98	1.1428	0.5413	1.1425	1.0689
No log	50.0	100	0.7863	0.6209	0.7860	0.8865
No log	51.0	102	0.6395	0.6730	0.6391	0.7995
No log	52.0	104	0.7538	0.6266	0.7534	0.8680

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0