BERT_V8_sp10_lw40_ex100_lo50_k10_k10_fold4

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: 0.5427
• Qwk: 0.5669
• Mse: 0.5427
• Rmse: 0.7367

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 OptimizerNames.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	5	6.7078	0.0	6.7078	2.5900
No log	2.0	10	4.4067	0.0079	4.4067	2.0992
No log	3.0	15	2.4365	0.0175	2.4365	1.5609
No log	4.0	20	1.3072	0.0107	1.3072	1.1433
No log	5.0	25	0.8520	0.3982	0.8520	0.9230
No log	6.0	30	0.6719	0.3895	0.6719	0.8197
No log	7.0	35	0.9309	0.2902	0.9309	0.9648
No log	8.0	40	0.7205	0.4966	0.7205	0.8488
No log	9.0	45	0.7712	0.4634	0.7712	0.8782
No log	10.0	50	0.8108	0.4906	0.8108	0.9004
No log	11.0	55	0.5563	0.5587	0.5563	0.7458
No log	12.0	60	0.5058	0.5962	0.5058	0.7112
No log	13.0	65	0.5596	0.6458	0.5596	0.7481
No log	14.0	70	0.5800	0.6231	0.5800	0.7616
No log	15.0	75	0.5088	0.5849	0.5088	0.7133
No log	16.0	80	0.5191	0.6323	0.5191	0.7205
No log	17.0	85	0.5390	0.5711	0.5390	0.7342
No log	18.0	90	0.5895	0.6454	0.5895	0.7678
No log	19.0	95	0.5398	0.6112	0.5398	0.7347
No log	20.0	100	0.5523	0.5777	0.5523	0.7432
No log	21.0	105	0.7372	0.5103	0.7372	0.8586
No log	22.0	110	0.6965	0.5279	0.6965	0.8346
No log	23.0	115	0.5263	0.5886	0.5263	0.7255
No log	24.0	120	0.5104	0.5909	0.5104	0.7144
No log	25.0	125	0.5223	0.5781	0.5223	0.7227
No log	26.0	130	0.5991	0.5468	0.5991	0.7740
No log	27.0	135	0.5744	0.5574	0.5744	0.7579
No log	28.0	140	0.5720	0.5672	0.5720	0.7563
No log	29.0	145	0.5213	0.5593	0.5213	0.7220
No log	30.0	150	0.6727	0.5252	0.6727	0.8202
No log	31.0	155	0.5432	0.5692	0.5432	0.7370
No log	32.0	160	0.5245	0.5905	0.5245	0.7242
No log	33.0	165	0.5201	0.5338	0.5201	0.7212
No log	34.0	170	0.5244	0.5561	0.5244	0.7242
No log	35.0	175	0.5202	0.5556	0.5202	0.7212
No log	36.0	180	0.5320	0.5544	0.5320	0.7294
No log	37.0	185	0.5401	0.5909	0.5401	0.7349
No log	38.0	190	0.6913	0.5194	0.6913	0.8314
No log	39.0	195	0.5447	0.5519	0.5447	0.7380
No log	40.0	200	0.5087	0.5540	0.5087	0.7132
No log	41.0	205	0.5323	0.5580	0.5323	0.7296
No log	42.0	210	0.5400	0.5569	0.5400	0.7349
No log	43.0	215	0.5427	0.5669	0.5427	0.7367

Framework versions

• Transformers 4.51.1
• Pytorch 2.5.1+cu124
• Datasets 3.5.0
• Tokenizers 0.21.0