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	2	8.3229	0.0	8.3235	2.8850
No log	2.0	4	5.4697	0.0305	5.4701	2.3388
No log	3.0	6	4.3691	0.0074	4.3696	2.0904
No log	4.0	8	3.3312	0.0	3.3316	1.8253
No log	5.0	10	2.7022	0.0	2.7025	1.6439
No log	6.0	12	2.0488	0.0539	2.0492	1.4315
No log	7.0	14	1.6058	0.0213	1.6063	1.2674
No log	8.0	16	1.2831	0.0107	1.2835	1.1329
No log	9.0	18	1.0581	0.0	1.0586	1.0289
No log	10.0	20	0.9135	0.0174	0.9139	0.9560
No log	11.0	22	0.7748	0.4326	0.7751	0.8804
No log	12.0	24	0.6821	0.4441	0.6824	0.8261
No log	13.0	26	0.6655	0.5006	0.6657	0.8159
No log	14.0	28	0.9482	0.4392	0.9486	0.9739
No log	15.0	30	0.6522	0.5299	0.6525	0.8078
No log	16.0	32	0.6068	0.5334	0.6070	0.7791
No log	17.0	34	0.7950	0.4575	0.7953	0.8918
No log	18.0	36	0.5313	0.5396	0.5314	0.7290
No log	19.0	38	0.5333	0.5141	0.5334	0.7304
No log	20.0	40	0.7438	0.4563	0.7441	0.8626
No log	21.0	42	0.5033	0.4778	0.5035	0.7095
No log	22.0	44	0.5222	0.4703	0.5224	0.7228
No log	23.0	46	1.0906	0.3540	1.0910	1.0445
No log	24.0	48	1.0160	0.3434	1.0163	1.0081
No log	25.0	50	0.6270	0.4850	0.6270	0.7918
No log	26.0	52	0.7566	0.4156	0.7565	0.8697
No log	27.0	54	0.6138	0.4984	0.6140	0.7836
No log	28.0	56	0.7150	0.4678	0.7153	0.8457
No log	29.0	58	0.5965	0.5267	0.5965	0.7723
No log	30.0	60	0.6063	0.5209	0.6063	0.7787
No log	31.0	62	0.7625	0.4666	0.7626	0.8733
No log	32.0	64	0.6807	0.4773	0.6808	0.8251
No log	33.0	66	0.6121	0.5192	0.6120	0.7823
No log	34.0	68	0.6670	0.5007	0.6670	0.8167
No log	35.0	70	0.7233	0.4725	0.7232	0.8504
No log	36.0	72	0.6248	0.5015	0.6247	0.7904
No log	37.0	74	0.6208	0.5281	0.6206	0.7878
No log	38.0	76	0.6634	0.4872	0.6632	0.8144
No log	39.0	78	0.6275	0.5259	0.6272	0.7920
No log	40.0	80	0.7262	0.4741	0.7260	0.8521
No log	41.0	82	0.5982	0.5336	0.5979	0.7733
No log	42.0	84	0.6655	0.5016	0.6654	0.8157
No log	43.0	86	0.6379	0.5292	0.6378	0.7986
No log	44.0	88	0.5746	0.5601	0.5744	0.7579
No log	45.0	90	0.6201	0.5274	0.6199	0.7873
No log	46.0	92	0.6396	0.5120	0.6394	0.7996
No log	47.0	94	0.6141	0.5267	0.6140	0.7836
No log	48.0	96	0.5704	0.5480	0.5702	0.7551
No log	49.0	98	0.5757	0.5576	0.5756	0.7587
No log	50.0	100	0.6468	0.5177	0.6467	0.8042
No log	51.0	102	0.5787	0.5710	0.5785	0.7606
No log	52.0	104	0.5930	0.5193	0.5928	0.7699
No log	53.0	106	0.7259	0.4415	0.7258	0.8519
No log	54.0	108	0.5968	0.5022	0.5967	0.7724
No log	55.0	110	0.5875	0.5239	0.5873	0.7664
No log	56.0	112	0.6424	0.4825	0.6424	0.8015
No log	57.0	114	0.5809	0.5258	0.5808	0.7621
No log	58.0	116	0.5899	0.5185	0.5897	0.7679
No log	59.0	118	0.7655	0.4309	0.7654	0.8749
No log	60.0	120	0.8690	0.4054	0.8689	0.9321
No log	61.0	122	0.6501	0.4709	0.6499	0.8062
No log	62.0	124	0.6615	0.5371	0.6612	0.8132
No log	63.0	126	0.6332	0.5361	0.6329	0.7956
No log	64.0	128	0.6534	0.49	0.6532	0.8082
No log	65.0	130	0.7385	0.4285	0.7384	0.8593
No log	66.0	132	0.6379	0.4811	0.6378	0.7986
No log	67.0	134	0.5806	0.5068	0.5805	0.7619
No log	68.0	136	0.6249	0.4913	0.6248	0.7904
No log	69.0	138	0.6241	0.5038	0.6240	0.7900
No log	70.0	140	0.5628	0.5256	0.5626	0.7501
No log	71.0	142	0.5642	0.5312	0.5640	0.7510
No log	72.0	144	0.6349	0.5143	0.6348	0.7967
No log	73.0	146	0.6285	0.5277	0.6284	0.7927
No log	74.0	148	0.5921	0.528	0.5920	0.7694
No log	75.0	150	0.6006	0.5460	0.6005	0.7749
No log	76.0	152	0.6056	0.5307	0.6055	0.7781
No log	77.0	154	0.5784	0.5207	0.5782	0.7604
No log	78.0	156	0.5950	0.5274	0.5949	0.7713
No log	79.0	158	0.5774	0.5276	0.5773	0.7598
No log	80.0	160	0.5860	0.5138	0.5859	0.7655
No log	81.0	162	0.6238	0.4855	0.6237	0.7897

Framework versions

Transformers 4.51.1
Pytorch 2.5.1+cu124
Datasets 3.5.0
Tokenizers 0.21.0

genki10/BERT_V8_sp10_lw40_ex50_lo100_k2_k2_fold2

BERT_V8_sp10_lw40_ex50_lo100_k2_k2_fold2

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

genki10/
BERT_V8_sp10_lw40_ex50_lo100_k2_k2_fold2