Version12NewTestASAP_FineTuningBERT_AugV12_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: 0.5162
• Qwk: 0.5854
• Mse: 0.5162
• Rmse: 0.7185

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	3	7.7464	0.0	7.7464	2.7832
No log	2.0	6	5.3347	0.0093	5.3347	2.3097
No log	3.0	9	3.6844	0.0	3.6844	1.9195
No log	4.0	12	2.7343	0.0	2.7343	1.6536
No log	5.0	15	1.8603	0.1096	1.8603	1.3639
No log	6.0	18	1.5129	0.0747	1.5129	1.2300
No log	7.0	21	1.4817	0.0747	1.4817	1.2173
No log	8.0	24	1.2973	0.0	1.2973	1.1390
No log	9.0	27	1.1178	0.0	1.1178	1.0573
No log	10.0	30	1.1232	0.1012	1.1232	1.0598
No log	11.0	33	1.0465	0.1724	1.0465	1.0230
No log	12.0	36	0.9349	0.3007	0.9349	0.9669
No log	13.0	39	0.8622	0.3970	0.8622	0.9285
No log	14.0	42	0.8531	0.4677	0.8531	0.9236
No log	15.0	45	0.5885	0.5318	0.5885	0.7671
No log	16.0	48	1.0167	0.4615	1.0167	1.0083
No log	17.0	51	0.4888	0.5890	0.4888	0.6992
No log	18.0	54	0.7100	0.5441	0.7100	0.8426
No log	19.0	57	0.4998	0.5528	0.4998	0.7070
No log	20.0	60	0.8361	0.4741	0.8361	0.9144
No log	21.0	63	0.5561	0.5105	0.5561	0.7457
No log	22.0	66	0.7953	0.4848	0.7953	0.8918
No log	23.0	69	0.5596	0.5633	0.5596	0.7481
No log	24.0	72	0.5753	0.5740	0.5753	0.7585
No log	25.0	75	0.6451	0.5604	0.6451	0.8032
No log	26.0	78	0.6664	0.5342	0.6664	0.8163
No log	27.0	81	0.6055	0.5710	0.6055	0.7781
No log	28.0	84	0.6412	0.5468	0.6412	0.8007
No log	29.0	87	0.6285	0.5401	0.6285	0.7928
No log	30.0	90	0.7630	0.4609	0.7630	0.8735
No log	31.0	93	0.5766	0.5050	0.5766	0.7593
No log	32.0	96	0.6521	0.5238	0.6521	0.8075
No log	33.0	99	0.5810	0.5719	0.5810	0.7622
No log	34.0	102	0.5525	0.5960	0.5525	0.7433
No log	35.0	105	0.5137	0.5955	0.5137	0.7167
No log	36.0	108	0.5492	0.5753	0.5492	0.7411
No log	37.0	111	0.6113	0.5877	0.6113	0.7819
No log	38.0	114	0.5622	0.5705	0.5622	0.7498
No log	39.0	117	0.5749	0.5364	0.5749	0.7582
No log	40.0	120	0.6364	0.5295	0.6364	0.7977
No log	41.0	123	0.6428	0.5663	0.6428	0.8017
No log	42.0	126	0.5573	0.4909	0.5573	0.7466
No log	43.0	129	0.5946	0.5543	0.5946	0.7711
No log	44.0	132	0.6146	0.5482	0.6146	0.7840
No log	45.0	135	0.5170	0.5977	0.5170	0.7190
No log	46.0	138	0.7351	0.5571	0.7351	0.8574
No log	47.0	141	0.5230	0.6141	0.5230	0.7232
No log	48.0	144	0.5870	0.5723	0.5870	0.7661
No log	49.0	147	0.5428	0.5222	0.5428	0.7367
No log	50.0	150	0.6000	0.5008	0.6000	0.7746
No log	51.0	153	0.5668	0.5280	0.5668	0.7529
No log	52.0	156	0.5792	0.5068	0.5792	0.7611
No log	53.0	159	0.6189	0.5581	0.6189	0.7867
No log	54.0	162	0.5480	0.5018	0.5480	0.7402
No log	55.0	165	0.5632	0.5315	0.5632	0.7504
No log	56.0	168	0.5869	0.5423	0.5869	0.7661
No log	57.0	171	0.5428	0.5520	0.5428	0.7367
No log	58.0	174	0.5772	0.5210	0.5772	0.7597
No log	59.0	177	0.5412	0.5392	0.5412	0.7356
No log	60.0	180	0.5827	0.5380	0.5827	0.7633
No log	61.0	183	0.5617	0.5068	0.5617	0.7495
No log	62.0	186	0.5310	0.5596	0.5310	0.7287
No log	63.0	189	0.5745	0.5710	0.5745	0.7580
No log	64.0	192	0.5142	0.5777	0.5142	0.7171
No log	65.0	195	0.5310	0.5736	0.5310	0.7287
No log	66.0	198	0.5945	0.5879	0.5945	0.7711
No log	67.0	201	0.5162	0.5854	0.5162	0.7185

Framework versions

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