FineTuning_Method_2_SC

This model is a fine-tuned version of rafsankabir/Pretrained_E13_Method2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.3223
• Accuracy: 0.6790
• F1 Macro: 0.6487

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 40
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Macro
No log	0.32	500	1.0745	0.3976	0.1896
1.0543	0.64	1000	0.9059	0.5967	0.4614
1.0543	0.95	1500	0.8259	0.6414	0.5633
0.8389	1.27	2000	0.8177	0.6394	0.5715
0.8389	1.59	2500	0.8269	0.6356	0.5724
0.7713	1.91	3000	0.7916	0.6631	0.6238
0.7713	2.23	3500	0.7996	0.6745	0.6155
0.6734	2.54	4000	0.7921	0.6624	0.6307
0.6734	2.86	4500	0.7743	0.6726	0.6459
0.6309	3.18	5000	0.8343	0.6803	0.6382
0.6309	3.5	5500	0.8233	0.6784	0.6390
0.5582	3.82	6000	0.8678	0.6631	0.6273
0.5582	4.13	6500	0.8621	0.6758	0.6368
0.4988	4.45	7000	0.9389	0.6720	0.6386
0.4988	4.77	7500	0.9067	0.6918	0.6505
0.4885	5.09	8000	0.9116	0.6937	0.6583
0.4885	5.41	8500	1.0357	0.6822	0.6459
0.427	5.73	9000	0.9428	0.6847	0.6479
0.427	6.04	9500	1.0233	0.6752	0.6531
0.4034	6.36	10000	1.1578	0.6835	0.6515
0.4034	6.68	10500	1.1870	0.6790	0.6545
0.4053	7.0	11000	1.0370	0.7007	0.6651
0.4053	7.32	11500	1.2087	0.6822	0.6497
0.3545	7.63	12000	1.2255	0.6847	0.6605
0.3545	7.95	12500	1.2710	0.6905	0.6609
0.3437	8.27	13000	1.3646	0.6918	0.6618
0.3437	8.59	13500	1.3767	0.6879	0.6563
0.3407	8.91	14000	1.2705	0.6796	0.6506
0.3407	9.22	14500	1.4605	0.6803	0.6496
0.2876	9.54	15000	1.4202	0.6860	0.6555
0.2876	9.86	15500	1.4151	0.6847	0.6517
0.3035	10.18	16000	1.4536	0.6713	0.6514
0.3035	10.5	16500	1.4806	0.6828	0.6469
0.2733	10.81	17000	1.4596	0.6899	0.6552
0.2733	11.13	17500	1.6183	0.6886	0.6557
0.2562	11.45	18000	1.6054	0.6771	0.6591
0.2562	11.77	18500	1.5966	0.6701	0.6503
0.2582	12.09	19000	1.5659	0.6822	0.6531
0.2582	12.4	19500	1.6146	0.6867	0.6575
0.2368	12.72	20000	1.6207	0.6899	0.6629
0.2368	13.04	20500	1.5220	0.6918	0.6640
0.245	13.36	21000	1.6572	0.6720	0.6489
0.245	13.68	21500	1.6443	0.6860	0.6590
0.2226	13.99	22000	1.6238	0.6847	0.6589
0.2226	14.31	22500	1.7241	0.6777	0.6521
0.2117	14.63	23000	1.6134	0.6867	0.6580
0.2117	14.95	23500	1.6723	0.6911	0.6618
0.2056	15.27	24000	1.6257	0.6892	0.6529
0.2056	15.59	24500	1.7072	0.6796	0.6531
0.1859	15.9	25000	1.7174	0.6771	0.6554
0.1859	16.22	25500	1.6951	0.6879	0.6555
0.1725	16.54	26000	1.7240	0.6905	0.6632
0.1725	16.86	26500	1.7126	0.6879	0.6608
0.1817	17.18	27000	1.7949	0.6847	0.6520
0.1817	17.49	27500	1.7694	0.6911	0.6622
0.1617	17.81	28000	1.7891	0.6828	0.6527
0.1617	18.13	28500	1.7860	0.6790	0.6526
0.1628	18.45	29000	1.8127	0.6867	0.6605
0.1628	18.77	29500	1.7317	0.6892	0.6610
0.1736	19.08	30000	1.7273	0.6899	0.6569
0.1736	19.4	30500	1.7853	0.6854	0.6584
0.1441	19.72	31000	1.7866	0.6918	0.6624
0.1441	20.04	31500	1.7842	0.6873	0.6580
0.1392	20.36	32000	1.8669	0.6860	0.6597
0.1392	20.67	32500	1.8392	0.6899	0.6639
0.159	20.99	33000	1.8412	0.6784	0.6552
0.159	21.31	33500	1.8673	0.6854	0.6584
0.1275	21.63	34000	1.8622	0.6854	0.6571
0.1275	21.95	34500	1.8622	0.6796	0.6583
0.1216	22.26	35000	1.9509	0.6854	0.6604
0.1216	22.58	35500	1.9425	0.6809	0.6550
0.1351	22.9	36000	1.9496	0.6784	0.6559
0.1351	23.22	36500	1.9685	0.6847	0.6582
0.1221	23.54	37000	1.9112	0.6911	0.6642
0.1221	23.85	37500	1.9341	0.6726	0.6526
0.1155	24.17	38000	1.9573	0.6899	0.6614
0.1155	24.49	38500	1.9853	0.6873	0.6580
0.1139	24.81	39000	1.9915	0.6790	0.6533
0.1139	25.13	39500	1.9997	0.6796	0.6539
0.1166	25.45	40000	1.9994	0.6847	0.6592
0.1166	25.76	40500	1.9848	0.6745	0.6513
0.1128	26.08	41000	2.0095	0.6867	0.6578
0.1128	26.4	41500	2.0585	0.6822	0.6547
0.1048	26.72	42000	2.0293	0.6777	0.6510
0.1048	27.04	42500	2.0797	0.6758	0.6512
0.1	27.35	43000	2.1162	0.6822	0.6544
0.1	27.67	43500	2.0569	0.6835	0.6538
0.1106	27.99	44000	2.0991	0.6828	0.6565
0.1106	28.31	44500	2.0976	0.6841	0.6563
0.0886	28.63	45000	2.1305	0.6854	0.6532
0.0886	28.94	45500	2.1015	0.6867	0.6564
0.1027	29.26	46000	2.1105	0.6867	0.6559
0.1027	29.58	46500	2.1396	0.6765	0.6499
0.1057	29.9	47000	2.1237	0.6790	0.6501
0.1057	30.22	47500	2.1849	0.6790	0.6518
0.0876	30.53	48000	2.1346	0.6841	0.6533
0.0876	30.85	48500	2.1441	0.6828	0.6540
0.0856	31.17	49000	2.1528	0.6911	0.6600
0.0856	31.49	49500	2.1725	0.6847	0.6509
0.0869	31.81	50000	2.2085	0.6771	0.6503
0.0869	32.12	50500	2.2606	0.6688	0.6434
0.0848	32.44	51000	2.2510	0.6745	0.6451
0.0848	32.76	51500	2.2528	0.6739	0.6496
0.0816	33.08	52000	2.2532	0.6758	0.6503
0.0816	33.4	52500	2.2356	0.6803	0.6500
0.0793	33.72	53000	2.2579	0.6745	0.6483
0.0793	34.03	53500	2.2126	0.6816	0.6520
0.0767	34.35	54000	2.2504	0.6803	0.6497
0.0767	34.67	54500	2.2601	0.6803	0.6524
0.0844	34.99	55000	2.2785	0.6733	0.6470
0.0844	35.31	55500	2.2756	0.6784	0.6520
0.0755	35.62	56000	2.2813	0.6816	0.6542
0.0755	35.94	56500	2.2752	0.6803	0.6518
0.077	36.26	57000	2.2815	0.6796	0.6518
0.077	36.58	57500	2.2861	0.6803	0.6514
0.0752	36.9	58000	2.2929	0.6771	0.6505
0.0752	37.21	58500	2.2859	0.6816	0.6537
0.0698	37.53	59000	2.3117	0.6796	0.6525
0.0698	37.85	59500	2.3038	0.6816	0.6511
0.0613	38.17	60000	2.3176	0.6765	0.6477
0.0613	38.49	60500	2.3131	0.6796	0.6493
0.0706	38.8	61000	2.3161	0.6777	0.6477
0.0706	39.12	61500	2.3127	0.6784	0.6484
0.0678	39.44	62000	2.3174	0.6765	0.6467
0.0678	39.76	62500	2.3223	0.6790	0.6487

Framework versions

• Transformers 4.29.2
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3