Version2ASAP_FineTuningBERT_AugV8_k3_task1_organization_k3_k3_fold3

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: 0.5892
• Qwk: 0.5774
• Mse: 0.5895
• Rmse: 0.7678

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	9.4987	0.0	9.4970	3.0817
No log	2.0	6	6.8762	0.0	6.8749	2.6220
No log	3.0	9	4.8867	0.0093	4.8855	2.2103
No log	4.0	12	3.4657	0.0	3.4647	1.8614
No log	5.0	15	2.5127	0.0980	2.5119	1.5849
No log	6.0	18	1.8914	0.0202	1.8906	1.3750
No log	7.0	21	1.3800	0.0202	1.3794	1.1745
No log	8.0	24	1.1948	0.0202	1.1941	1.0927
No log	9.0	27	2.0060	0.1502	2.0053	1.4161
No log	10.0	30	1.1674	0.0365	1.1670	1.0803
No log	11.0	33	1.0969	0.0202	1.0962	1.0470
No log	12.0	36	1.1523	0.0365	1.1518	1.0732
No log	13.0	39	0.8589	0.2972	0.8587	0.9267
No log	14.0	42	0.8938	0.2983	0.8938	0.9454
No log	15.0	45	0.7197	0.4321	0.7196	0.8483
No log	16.0	48	0.6540	0.4354	0.6543	0.8089
No log	17.0	51	0.6279	0.3700	0.6278	0.7923
No log	18.0	54	0.5813	0.4549	0.5815	0.7625
No log	19.0	57	0.6977	0.3382	0.6977	0.8353
No log	20.0	60	0.6340	0.4092	0.6342	0.7963
No log	21.0	63	0.5500	0.4472	0.5501	0.7417
No log	22.0	66	0.5115	0.5090	0.5116	0.7152
No log	23.0	69	0.5381	0.5605	0.5381	0.7336
No log	24.0	72	0.5738	0.5666	0.5741	0.7577
No log	25.0	75	0.5425	0.5831	0.5426	0.7366
No log	26.0	78	0.6025	0.5860	0.6027	0.7763
No log	27.0	81	0.6201	0.5772	0.6203	0.7876
No log	28.0	84	0.8224	0.5279	0.8227	0.9070
No log	29.0	87	0.6582	0.5783	0.6587	0.8116
No log	30.0	90	0.6277	0.5904	0.6280	0.7925
No log	31.0	93	0.6157	0.5810	0.6160	0.7848
No log	32.0	96	0.6658	0.5688	0.6662	0.8162
No log	33.0	99	0.7505	0.5345	0.7511	0.8666
No log	34.0	102	0.7052	0.5377	0.7057	0.8400
No log	35.0	105	0.6072	0.5791	0.6076	0.7795
No log	36.0	108	0.6696	0.5643	0.6700	0.8185
No log	37.0	111	0.6451	0.5888	0.6456	0.8035
No log	38.0	114	0.6244	0.5750	0.6248	0.7904
No log	39.0	117	0.6714	0.5371	0.6717	0.8196
No log	40.0	120	0.6318	0.5607	0.6322	0.7951
No log	41.0	123	0.7037	0.5271	0.7040	0.8390
No log	42.0	126	0.6035	0.5694	0.6040	0.7772
No log	43.0	129	0.7107	0.5383	0.7111	0.8433
No log	44.0	132	0.6464	0.5463	0.6468	0.8042
No log	45.0	135	0.7366	0.5338	0.7369	0.8584
No log	46.0	138	0.5861	0.5803	0.5865	0.7658
No log	47.0	141	0.6159	0.5654	0.6162	0.7850
No log	48.0	144	0.6227	0.5632	0.6231	0.7893
No log	49.0	147	0.6299	0.5653	0.6302	0.7938
No log	50.0	150	0.5892	0.5774	0.5895	0.7678

Framework versions

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