Version2ASAP_FineTuningBERT_AugV6_k3_task1_organization_k3_k3_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: 0.8196
• Qwk: 0.4063
• Mse: 0.8192
• Rmse: 0.9051

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	6	7.0457	0.0	7.0458	2.6544
No log	2.0	12	3.5401	0.0039	3.5405	1.8816
No log	3.0	18	1.6435	0.0475	1.6440	1.2822
No log	4.0	24	0.9649	0.0345	0.9653	0.9825
No log	5.0	30	1.4865	0.1092	1.4869	1.2194
No log	6.0	36	0.7925	0.4008	0.7927	0.8903
No log	7.0	42	1.0168	0.2895	1.0168	1.0084
No log	8.0	48	1.0964	0.1592	1.0964	1.0471
No log	9.0	54	0.6104	0.3835	0.6104	0.7813
No log	10.0	60	0.7211	0.3082	0.7211	0.8492
No log	11.0	66	0.9918	0.2451	0.9915	0.9957
No log	12.0	72	0.7481	0.3640	0.7477	0.8647
No log	13.0	78	0.8308	0.4197	0.8304	0.9113
No log	14.0	84	1.1159	0.2925	1.1156	1.0562
No log	15.0	90	1.2154	0.3002	1.2149	1.1022
No log	16.0	96	0.9392	0.4070	0.9387	0.9689
No log	17.0	102	0.9326	0.3782	0.9321	0.9655
No log	18.0	108	0.9293	0.3637	0.9288	0.9637
No log	19.0	114	0.8598	0.4094	0.8592	0.9269
No log	20.0	120	0.7120	0.4885	0.7113	0.8434
No log	21.0	126	0.7631	0.4662	0.7626	0.8733
No log	22.0	132	0.8345	0.3949	0.8341	0.9133
No log	23.0	138	0.8146	0.4097	0.8142	0.9023
No log	24.0	144	0.7949	0.4205	0.7943	0.8912
No log	25.0	150	0.8530	0.4165	0.8525	0.9233
No log	26.0	156	0.8076	0.4655	0.8070	0.8983
No log	27.0	162	0.7263	0.4682	0.7258	0.8519
No log	28.0	168	0.7715	0.4346	0.7710	0.8781
No log	29.0	174	0.7572	0.4381	0.7568	0.8699
No log	30.0	180	0.8424	0.3874	0.8420	0.9176
No log	31.0	186	0.7832	0.4171	0.7829	0.8848
No log	32.0	192	0.8584	0.3981	0.8580	0.9263
No log	33.0	198	0.8111	0.4253	0.8107	0.9004
No log	34.0	204	0.8412	0.3733	0.8408	0.9169
No log	35.0	210	0.7940	0.3985	0.7935	0.8908
No log	36.0	216	0.7850	0.4319	0.7845	0.8857
No log	37.0	222	0.8200	0.3870	0.8195	0.9052
No log	38.0	228	0.9315	0.3338	0.9312	0.9650
No log	39.0	234	0.8157	0.3980	0.8154	0.9030
No log	40.0	240	0.8196	0.4063	0.8192	0.9051

Framework versions

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