Version_weird_ASAP_FineTuningBERT_AugV12_k10_task1_organization_k10_k10_fold2

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: 0.8028
• Qwk: 0.5731
• Mse: 0.8018
• Rmse: 0.8954

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	12.2952	0.0014	12.2952	3.5064
No log	2.0	6	9.1980	0.0	9.1982	3.0328
No log	3.0	9	7.4767	0.0	7.4767	2.7344
No log	4.0	12	4.9672	0.0227	4.9676	2.2288
No log	5.0	15	4.1565	0.0	4.1567	2.0388
No log	6.0	18	2.2339	0.1188	2.2343	1.4948
No log	7.0	21	1.4408	0.0213	1.4413	1.2005
No log	8.0	24	1.0764	0.0107	1.0768	1.0377
No log	9.0	27	0.8391	0.3043	0.8395	0.9162
No log	10.0	30	0.8339	0.0430	0.8344	0.9135
No log	11.0	33	0.9497	0.0327	0.9501	0.9748
No log	12.0	36	1.3084	0.0045	1.3089	1.1441
No log	13.0	39	1.6864	0.0596	1.6867	1.2987
No log	14.0	42	1.9860	0.0394	1.9862	1.4093
No log	15.0	45	2.2525	0.0660	2.2522	1.5007
No log	16.0	48	1.9149	0.2144	1.9140	1.3835
No log	17.0	51	0.7209	0.5306	0.7201	0.8486
No log	18.0	54	0.9702	0.5485	0.9690	0.9844
No log	19.0	57	0.9280	0.5405	0.9268	0.9627
No log	20.0	60	0.9608	0.5708	0.9594	0.9795
No log	21.0	63	0.5862	0.6151	0.5856	0.7652
No log	22.0	66	1.2980	0.4754	1.2964	1.1386
No log	23.0	69	1.2148	0.4755	1.2134	1.1015
No log	24.0	72	0.5068	0.6207	0.5062	0.7114
No log	25.0	75	0.8001	0.5837	0.7990	0.8939
No log	26.0	78	0.5592	0.6342	0.5586	0.7474
No log	27.0	81	0.8511	0.5480	0.8500	0.9220
No log	28.0	84	0.5569	0.6283	0.5563	0.7458
No log	29.0	87	1.0873	0.5054	1.0862	1.0422
No log	30.0	90	1.0732	0.5105	1.0721	1.0354
No log	31.0	93	0.6005	0.6270	0.6000	0.7746
No log	32.0	96	0.7228	0.6123	0.7220	0.8497
No log	33.0	99	1.4819	0.4786	1.4805	1.2167
No log	34.0	102	0.9616	0.5356	0.9605	0.9800
No log	35.0	105	0.5874	0.6232	0.5868	0.7660
No log	36.0	108	0.6780	0.6192	0.6772	0.8229
No log	37.0	111	0.8785	0.5894	0.8772	0.9366
No log	38.0	114	0.7288	0.5846	0.7283	0.8534
No log	39.0	117	0.6293	0.6072	0.6289	0.7930
No log	40.0	120	0.8761	0.5375	0.8750	0.9354
No log	41.0	123	0.8686	0.5431	0.8676	0.9315
No log	42.0	126	0.6751	0.6416	0.6744	0.8212
No log	43.0	129	0.7368	0.6340	0.7360	0.8579
No log	44.0	132	1.0939	0.5327	1.0925	1.0452
No log	45.0	135	1.0294	0.52	1.0282	1.0140
No log	46.0	138	0.6799	0.5976	0.6790	0.8240
No log	47.0	141	0.8283	0.5705	0.8273	0.9095
No log	48.0	144	0.7470	0.6135	0.7461	0.8638
No log	49.0	147	0.7234	0.6247	0.7226	0.8501
No log	50.0	150	0.8410	0.5769	0.8400	0.9165
No log	51.0	153	0.7409	0.5779	0.7400	0.8603
No log	52.0	156	0.9363	0.5255	0.9353	0.9671
No log	53.0	159	0.9195	0.5409	0.9184	0.9583
No log	54.0	162	0.7369	0.5935	0.7360	0.8579
No log	55.0	165	0.6877	0.6308	0.6869	0.8288
No log	56.0	168	0.7305	0.6064	0.7296	0.8542
No log	57.0	171	0.9400	0.5486	0.9389	0.9690
No log	58.0	174	0.8382	0.5510	0.8372	0.9150
No log	59.0	177	0.6619	0.6133	0.6611	0.8131
No log	60.0	180	0.7234	0.5847	0.7225	0.8500
No log	61.0	183	0.8631	0.5590	0.8621	0.9285
No log	62.0	186	0.8028	0.5731	0.8018	0.8954

Framework versions

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