ASAP_FineTuningBERT_UnAugV5_k1_task1_organization_fold0

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: 0.6550
• Qwk: 0.6278
• Mse: 0.6550
• Rmse: 0.8094

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	2.0	2	10.5351	0.0061	10.5351	3.2458
No log	4.0	4	8.2198	0.0054	8.2198	2.8670
No log	6.0	6	6.3796	0.0029	6.3796	2.5258
No log	8.0	8	4.9785	0.0139	4.9785	2.2313
9.3005	10.0	10	3.7548	0.0115	3.7548	1.9377
9.3005	12.0	12	2.7153	0.0077	2.7153	1.6478
9.3005	14.0	14	1.8897	0.0520	1.8897	1.3747
9.3005	16.0	16	1.3712	0.0520	1.3712	1.1710
9.3005	18.0	18	1.0739	0.0520	1.0739	1.0363
3.4599	20.0	20	0.8739	0.0664	0.8739	0.9348
3.4599	22.0	22	0.7690	0.3702	0.7690	0.8769
3.4599	24.0	24	0.6434	0.4429	0.6434	0.8021
3.4599	26.0	26	0.6024	0.4529	0.6024	0.7761
3.4599	28.0	28	0.5723	0.4703	0.5723	0.7565
1.4411	30.0	30	0.5559	0.5397	0.5559	0.7456
1.4411	32.0	32	0.5948	0.4984	0.5948	0.7713
1.4411	34.0	34	0.5494	0.5483	0.5494	0.7412
1.4411	36.0	36	0.6670	0.5671	0.6670	0.8167
1.4411	38.0	38	0.5804	0.6215	0.5804	0.7618
0.6198	40.0	40	0.7679	0.5270	0.7679	0.8763
0.6198	42.0	42	0.6548	0.5980	0.6548	0.8092
0.6198	44.0	44	0.5776	0.6053	0.5776	0.7600
0.6198	46.0	46	1.0709	0.4333	1.0709	1.0349
0.6198	48.0	48	1.0767	0.4461	1.0767	1.0376
0.3311	50.0	50	0.6082	0.6208	0.6082	0.7799
0.3311	52.0	52	0.6632	0.6153	0.6632	0.8144
0.3311	54.0	54	0.8101	0.5604	0.8101	0.9000
0.3311	56.0	56	0.6507	0.6362	0.6507	0.8067
0.3311	58.0	58	0.7474	0.5957	0.7474	0.8645
0.1608	60.0	60	0.7510	0.5770	0.7510	0.8666
0.1608	62.0	62	0.6489	0.6265	0.6489	0.8056
0.1608	64.0	64	0.8040	0.5634	0.8040	0.8966
0.1608	66.0	66	0.6306	0.6342	0.6306	0.7941
0.1608	68.0	68	0.6491	0.6393	0.6491	0.8056
0.0968	70.0	70	0.8924	0.5310	0.8924	0.9447
0.0968	72.0	72	0.8359	0.5514	0.8359	0.9143
0.0968	74.0	74	0.6026	0.6239	0.6026	0.7763
0.0968	76.0	76	0.5893	0.6221	0.5893	0.7676
0.0968	78.0	78	0.6519	0.6201	0.6519	0.8074
0.073	80.0	80	0.9077	0.5460	0.9077	0.9527
0.073	82.0	82	0.9019	0.5462	0.9019	0.9497
0.073	84.0	84	0.7039	0.6093	0.7039	0.8390
0.073	86.0	86	0.6333	0.6300	0.6333	0.7958
0.073	88.0	88	0.6698	0.6237	0.6698	0.8184
0.0569	90.0	90	0.7536	0.5817	0.7536	0.8681
0.0569	92.0	92	0.8272	0.5526	0.8272	0.9095
0.0569	94.0	94	0.7834	0.5585	0.7834	0.8851
0.0569	96.0	96	0.7058	0.6116	0.7058	0.8401
0.0569	98.0	98	0.6605	0.6294	0.6605	0.8127
0.0526	100.0	100	0.6550	0.6278	0.6550	0.8094

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.2.0
• Tokenizers 0.19.1