Version22NewTestASAP_FineTuningBERT_AugV22_k3_task1_organization_k3_k3_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: 1.3531
• Qwk: 0.2634
• Mse: 1.3528
• Rmse: 1.1631

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	11.0443	-0.0097	11.0415	3.3229
No log	2.0	4	10.4552	0.0	10.4526	3.2330
No log	3.0	6	9.8887	0.0	9.8861	3.1442
No log	4.0	8	8.9520	0.0	8.9495	2.9916
No log	5.0	10	7.4046	0.0	7.4022	2.7207
No log	6.0	12	5.8302	0.0054	5.8279	2.4141
No log	7.0	14	4.5404	0.0040	4.5384	2.1303
No log	8.0	16	3.5596	0.0	3.5577	1.8862
No log	9.0	18	2.8863	0.0	2.8844	1.6984
No log	10.0	20	2.2986	0.1024	2.2968	1.5155
No log	11.0	22	1.8419	0.0211	1.8402	1.3565
No log	12.0	24	1.4991	0.0	1.4975	1.2237
No log	13.0	26	1.2627	0.0	1.2612	1.1230
No log	14.0	28	1.1146	0.0106	1.1133	1.0551
No log	15.0	30	1.0722	0.0828	1.0708	1.0348
No log	16.0	32	1.2178	0.1495	1.2163	1.1029
No log	17.0	34	1.0733	0.1689	1.0719	1.0353
No log	18.0	36	0.8999	0.1685	0.8986	0.9479
No log	19.0	38	1.0330	0.2085	1.0315	1.0156
No log	20.0	40	1.2705	0.1925	1.2689	1.1264
No log	21.0	42	0.8659	0.2653	0.8646	0.9299
No log	22.0	44	0.8910	0.2606	0.8898	0.9433
No log	23.0	46	1.1836	0.1996	1.1822	1.0873
No log	24.0	48	0.9485	0.2358	0.9473	0.9733
No log	25.0	50	0.9447	0.2685	0.9435	0.9714
No log	26.0	52	1.0033	0.2915	1.0021	1.0011
No log	27.0	54	0.8980	0.3647	0.8970	0.9471
No log	28.0	56	0.8033	0.4592	0.8024	0.8958
No log	29.0	58	1.2299	0.3317	1.2290	1.1086
No log	30.0	60	1.3553	0.2990	1.3544	1.1638
No log	31.0	62	0.7929	0.4942	0.7923	0.8901
No log	32.0	64	0.7631	0.4969	0.7625	0.8732
No log	33.0	66	1.1419	0.3692	1.1411	1.0682
No log	34.0	68	1.3416	0.3086	1.3407	1.1579
No log	35.0	70	0.8795	0.4094	0.8788	0.9374
No log	36.0	72	0.9278	0.4127	0.9271	0.9629
No log	37.0	74	1.2631	0.2894	1.2625	1.1236
No log	38.0	76	1.2406	0.2790	1.2400	1.1135
No log	39.0	78	1.1137	0.3448	1.1131	1.0550
No log	40.0	80	1.1942	0.2921	1.1936	1.0925
No log	41.0	82	1.2433	0.2763	1.2426	1.1147
No log	42.0	84	1.0703	0.3449	1.0697	1.0343
No log	43.0	86	1.0411	0.3591	1.0406	1.0201
No log	44.0	88	1.2113	0.2955	1.2107	1.1003
No log	45.0	90	1.2447	0.2873	1.2441	1.1154
No log	46.0	92	1.0769	0.3480	1.0764	1.0375
No log	47.0	94	1.1308	0.3261	1.1304	1.0632
No log	48.0	96	1.1927	0.2652	1.1922	1.0919
No log	49.0	98	1.2333	0.2628	1.2329	1.1104
No log	50.0	100	1.2276	0.3065	1.2272	1.1078
No log	51.0	102	1.2913	0.2746	1.2908	1.1362
No log	52.0	104	1.3032	0.2859	1.3027	1.1414
No log	53.0	106	1.2358	0.3345	1.2354	1.1115
No log	54.0	108	1.1625	0.3446	1.1622	1.0780
No log	55.0	110	1.1686	0.3233	1.1682	1.0808
No log	56.0	112	1.0841	0.3793	1.0838	1.0411
No log	57.0	114	1.0798	0.3726	1.0795	1.0390
No log	58.0	116	1.1177	0.3375	1.1174	1.0571
No log	59.0	118	1.1744	0.3566	1.1741	1.0836
No log	60.0	120	1.2571	0.3376	1.2569	1.1211
No log	61.0	122	1.3046	0.2570	1.3043	1.1421
No log	62.0	124	1.3531	0.2634	1.3528	1.1631

Framework versions

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