Version12LowDevASAP_FineTuningBERT_AugV12_k3_task1_organization_k3_k3_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: 1.1710
• Qwk: 0.4651
• Mse: 1.1710
• Rmse: 1.0821

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	1	13.0376	0.0	13.0376	3.6108
No log	2.0	2	11.3067	0.0130	11.3067	3.3625
No log	3.0	3	10.5137	0.0	10.5137	3.2425
No log	4.0	4	9.5737	0.0	9.5737	3.0941
No log	5.0	5	8.5732	0.0	8.5732	2.9280
No log	6.0	6	7.8321	0.0	7.8321	2.7986
No log	7.0	7	7.1757	0.0	7.1757	2.6788
No log	8.0	8	6.5251	0.0	6.5250	2.5544
No log	9.0	9	5.9631	0.0011	5.9631	2.4419
No log	10.0	10	5.3466	0.0	5.3466	2.3123
No log	11.0	11	4.7444	0.0	4.7444	2.1782
No log	12.0	12	4.3235	0.0	4.3235	2.0793
No log	13.0	13	3.8060	0.0	3.8060	1.9509
No log	14.0	14	3.3687	0.0	3.3687	1.8354
No log	15.0	15	3.1355	0.0	3.1355	1.7707
No log	16.0	16	2.6717	0.0591	2.6717	1.6345
No log	17.0	17	2.3149	0.0896	2.3149	1.5215
No log	18.0	18	2.0826	0.0370	2.0826	1.4431
No log	19.0	19	1.8430	0.0	1.8430	1.3576
No log	20.0	20	1.7412	0.0	1.7412	1.3196
No log	21.0	21	1.5365	0.0	1.5365	1.2396
No log	22.0	22	1.3490	0.0	1.3490	1.1615
No log	23.0	23	1.2545	0.0	1.2545	1.1200
No log	24.0	24	1.1513	0.0	1.1513	1.0730
No log	25.0	25	1.0470	0.0	1.0470	1.0232
No log	26.0	26	0.9826	-0.0132	0.9826	0.9913
No log	27.0	27	0.9245	0.2609	0.9245	0.9615
No log	28.0	28	0.8742	0.1935	0.8742	0.9350
No log	29.0	29	0.8354	0.1753	0.8354	0.9140
No log	30.0	30	0.8058	0.1753	0.8058	0.8977
No log	31.0	31	0.7779	0.2611	0.7779	0.8820
No log	32.0	32	0.7479	0.2233	0.7479	0.8648
No log	33.0	33	0.7215	0.2410	0.7215	0.8494
No log	34.0	34	0.7020	0.2689	0.7020	0.8379
No log	35.0	35	0.7051	0.2794	0.7051	0.8397
No log	36.0	36	0.7392	0.4023	0.7392	0.8598
No log	37.0	37	0.7276	0.3571	0.7276	0.8530
No log	38.0	38	0.6741	0.3595	0.6741	0.8210
No log	39.0	39	0.6671	0.3750	0.6671	0.8167
No log	40.0	40	0.6772	0.3907	0.6772	0.8229
No log	41.0	41	0.6767	0.4545	0.6767	0.8226
No log	42.0	42	0.6924	0.4595	0.6924	0.8321
No log	43.0	43	0.7227	0.4557	0.7227	0.8501
No log	44.0	44	0.7368	0.5196	0.7368	0.8584
No log	45.0	45	0.7520	0.5153	0.7520	0.8672
No log	46.0	46	0.7953	0.4924	0.7953	0.8918
No log	47.0	47	0.8220	0.4487	0.8220	0.9066
No log	48.0	48	0.8421	0.4583	0.8421	0.9177
No log	49.0	49	0.8786	0.4730	0.8786	0.9373
No log	50.0	50	0.9145	0.4776	0.9145	0.9563
No log	51.0	51	0.9550	0.4499	0.9550	0.9772
No log	52.0	52	1.0300	0.4256	1.0300	1.0149
No log	53.0	53	1.0563	0.4337	1.0563	1.0277
No log	54.0	54	0.9732	0.4937	0.9732	0.9865
No log	55.0	55	0.9658	0.4845	0.9658	0.9827
No log	56.0	56	1.0220	0.5038	1.0220	1.0109
No log	57.0	57	1.1144	0.4663	1.1144	1.0557
No log	58.0	58	1.0910	0.4740	1.0910	1.0445
No log	59.0	59	1.1251	0.4499	1.1251	1.0607
No log	60.0	60	1.2019	0.4221	1.2019	1.0963
No log	61.0	61	1.1440	0.4617	1.1440	1.0696
No log	62.0	62	1.0623	0.4863	1.0623	1.0307
No log	63.0	63	1.0592	0.5129	1.0592	1.0292
No log	64.0	64	1.1710	0.4651	1.1710	1.0821

Framework versions

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