ASAP_FineTuningBERT_AugV8_k7_task1_organization_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: 0.7978
• Qwk: 0.4989
• Mse: 0.7978
• Rmse: 0.8932

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	0.3333	2	9.7018	0.0037	9.7018	3.1148
No log	0.6667	4	9.2793	0.0018	9.2793	3.0462
No log	1.0	6	8.8818	0.0	8.8818	2.9802
No log	1.3333	8	8.2890	0.0	8.2890	2.8791
6.518	1.6667	10	7.3008	0.0	7.3008	2.7020
6.518	2.0	12	6.6691	0.0	6.6691	2.5825
6.518	2.3333	14	6.0001	0.0047	6.0001	2.4495
6.518	2.6667	16	5.0967	0.0128	5.0967	2.2576
6.518	3.0	18	4.2487	0.0040	4.2487	2.0612
3.7434	3.3333	20	3.5724	0.0040	3.5724	1.8901
3.7434	3.6667	22	2.9333	0.0040	2.9333	1.7127
3.7434	4.0	24	2.4539	0.0203	2.4539	1.5665
3.7434	4.3333	26	2.0124	0.0866	2.0124	1.4186
3.7434	4.6667	28	1.6811	0.0317	1.6811	1.2966
2.1748	5.0	30	1.4008	0.0213	1.4008	1.1835
2.1748	5.3333	32	1.3135	0.0213	1.3135	1.1461
2.1748	5.6667	34	1.3406	0.0317	1.3406	1.1579
2.1748	6.0	36	1.6241	0.0866	1.6241	1.2744
2.1748	6.3333	38	2.0719	0.1729	2.0719	1.4394
1.79	6.6667	40	1.6392	0.1281	1.6392	1.2803
1.79	7.0	42	1.2119	0.0675	1.2119	1.1009
1.79	7.3333	44	1.1311	0.0675	1.1311	1.0635
1.79	7.6667	46	1.3712	0.1144	1.3712	1.1710
1.79	8.0	48	1.5217	0.1612	1.5217	1.2336
1.586	8.3333	50	1.0776	0.1464	1.0776	1.0381
1.586	8.6667	52	0.9317	0.2521	0.9317	0.9652
1.586	9.0	54	1.4069	0.2006	1.4069	1.1861
1.586	9.3333	56	1.2012	0.2636	1.2012	1.0960
1.586	9.6667	58	1.0060	0.2730	1.0060	1.0030
1.2979	10.0	60	0.8809	0.3704	0.8809	0.9386
1.2979	10.3333	62	1.0739	0.3259	1.0739	1.0363
1.2979	10.6667	64	0.8681	0.3927	0.8681	0.9317
1.2979	11.0	66	0.7417	0.4617	0.7417	0.8612
1.2979	11.3333	68	0.7649	0.4523	0.7649	0.8746
0.7691	11.6667	70	0.7398	0.4705	0.7398	0.8601
0.7691	12.0	72	0.9612	0.4292	0.9612	0.9804
0.7691	12.3333	74	0.8471	0.4649	0.8471	0.9204
0.7691	12.6667	76	0.8901	0.4453	0.8901	0.9435
0.7691	13.0	78	0.9367	0.4327	0.9367	0.9678
0.4561	13.3333	80	0.8057	0.5252	0.8057	0.8976
0.4561	13.6667	82	0.8481	0.5055	0.8481	0.9209
0.4561	14.0	84	0.6920	0.5218	0.6920	0.8318
0.4561	14.3333	86	0.8464	0.4709	0.8464	0.9200
0.4561	14.6667	88	0.7155	0.5276	0.7155	0.8459
0.2563	15.0	90	0.6784	0.5192	0.6784	0.8236
0.2563	15.3333	92	0.7311	0.5202	0.7311	0.8551
0.2563	15.6667	94	0.7065	0.5422	0.7065	0.8405
0.2563	16.0	96	0.7486	0.4877	0.7486	0.8652
0.2563	16.3333	98	0.7347	0.5258	0.7347	0.8572
0.2127	16.6667	100	0.7380	0.5246	0.7380	0.8590
0.2127	17.0	102	0.7714	0.5287	0.7714	0.8783
0.2127	17.3333	104	0.7187	0.5695	0.7187	0.8478
0.2127	17.6667	106	0.8220	0.4896	0.8220	0.9066
0.2127	18.0	108	0.7005	0.5573	0.7005	0.8369
0.1651	18.3333	110	0.7453	0.5392	0.7453	0.8633
0.1651	18.6667	112	0.7390	0.5629	0.7390	0.8596
0.1651	19.0	114	0.7434	0.5224	0.7434	0.8622
0.1651	19.3333	116	0.7284	0.5356	0.7284	0.8535
0.1651	19.6667	118	0.7441	0.4967	0.7441	0.8626
0.1422	20.0	120	0.7752	0.4784	0.7752	0.8805
0.1422	20.3333	122	0.7277	0.5574	0.7277	0.8531
0.1422	20.6667	124	0.7978	0.4989	0.7978	0.8932

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0