haryoaw/scenario-TCR-xlmv-en-massive

scenario-TCR_data-en-massive_all_1_1

This model is a fine-tuned version of facebook/xlm-v-base on the massive dataset. It achieves the following results on the evaluation set:

• Loss: 2.3802
• Accuracy: 0.7101
• F1: 0.6551

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0.28	100	3.6542	0.0800	0.0085
No log	0.56	200	2.9766	0.3048	0.0953
No log	0.83	300	2.4835	0.3498	0.1168
No log	1.11	400	2.1305	0.4616	0.2154
2.7657	1.39	500	1.8889	0.5374	0.2791
2.7657	1.67	600	1.7326	0.5726	0.3208
2.7657	1.94	700	1.6536	0.5870	0.3726
2.7657	2.22	800	1.6709	0.5987	0.4014
2.7657	2.5	900	1.5460	0.6337	0.4720
1.1591	2.78	1000	1.5165	0.6434	0.4904
1.1591	3.06	1100	1.3861	0.6736	0.5237
1.1591	3.33	1200	1.3776	0.6739	0.5320
1.1591	3.61	1300	1.3753	0.6734	0.5521
1.1591	3.89	1400	1.4680	0.6624	0.5368
0.6194	4.17	1500	1.3899	0.6795	0.5520
0.6194	4.44	1600	1.5509	0.6640	0.5482
0.6194	4.72	1700	1.4034	0.6837	0.5764
0.6194	5.0	1800	1.4750	0.6739	0.5814
0.6194	5.28	1900	1.5321	0.6697	0.5761
0.3858	5.56	2000	1.5022	0.6822	0.5912
0.3858	5.83	2100	1.4612	0.6865	0.6016
0.3858	6.11	2200	1.4079	0.7034	0.6204
0.3858	6.39	2300	1.5165	0.6922	0.6296
0.3858	6.67	2400	1.6168	0.6736	0.6157
0.259	6.94	2500	1.5425	0.6948	0.6261
0.259	7.22	2600	1.6145	0.6796	0.6035
0.259	7.5	2700	1.5916	0.6824	0.6175
0.259	7.78	2800	1.5966	0.6977	0.6306
0.259	8.06	2900	1.4939	0.7125	0.6274
0.1759	8.33	3000	1.8425	0.6714	0.6170
0.1759	8.61	3100	1.6688	0.6923	0.6403
0.1759	8.89	3200	1.6218	0.6997	0.6220
0.1759	9.17	3300	1.7825	0.6829	0.6223
0.1759	9.44	3400	1.8706	0.6916	0.6294
0.1162	9.72	3500	1.8082	0.6884	0.6280
0.1162	10.0	3600	1.6708	0.7096	0.6338
0.1162	10.28	3700	1.7170	0.7100	0.6490
0.1162	10.56	3800	1.8575	0.6917	0.6264
0.1162	10.83	3900	1.8307	0.6959	0.6448
0.092	11.11	4000	1.9248	0.6958	0.6359
0.092	11.39	4100	1.7551	0.7162	0.6508
0.092	11.67	4200	1.8234	0.7072	0.6465
0.092	11.94	4300	2.1146	0.6790	0.6285
0.092	12.22	4400	1.9964	0.6909	0.6411
0.0582	12.5	4500	2.0290	0.6852	0.6313
0.0582	12.78	4600	2.0828	0.6838	0.6355
0.0582	13.06	4700	1.9272	0.7013	0.6312
0.0582	13.33	4800	1.9882	0.6959	0.6334
0.0582	13.61	4900	1.9552	0.7116	0.6511
0.0398	13.89	5000	2.0269	0.7060	0.6451
0.0398	14.17	5100	2.1377	0.6929	0.6414
0.0398	14.44	5200	2.1114	0.6880	0.6373
0.0398	14.72	5300	2.1517	0.6927	0.6438
0.0398	15.0	5400	2.2472	0.6921	0.6499
0.0311	15.28	5500	2.1801	0.6993	0.6557
0.0311	15.56	5600	2.1090	0.7020	0.6458
0.0311	15.83	5700	2.0049	0.7160	0.6590
0.0311	16.11	5800	2.2198	0.6959	0.6460
0.0311	16.39	5900	2.1074	0.7087	0.6519
0.0223	16.67	6000	2.0899	0.7096	0.6563
0.0223	16.94	6100	2.1736	0.7026	0.6546
0.0223	17.22	6200	2.1829	0.7004	0.6496
0.0223	17.5	6300	2.2041	0.6973	0.6450
0.0223	17.78	6400	2.1969	0.7074	0.6566
0.0178	18.06	6500	2.4021	0.6931	0.6515
0.0178	18.33	6600	2.2865	0.7092	0.6619
0.0178	18.61	6700	2.3086	0.7018	0.6504
0.0178	18.89	6800	2.2665	0.7054	0.6535
0.0178	19.17	6900	2.2723	0.7061	0.6525
0.0129	19.44	7000	2.2976	0.7030	0.6483
0.0129	19.72	7100	2.3634	0.7011	0.6514
0.0129	20.0	7200	2.3313	0.6971	0.6464
0.0129	20.28	7300	2.4373	0.6907	0.6439
0.0129	20.56	7400	2.2424	0.7139	0.6588
0.0125	20.83	7500	2.2329	0.7098	0.6547
0.0125	21.11	7600	2.2365	0.7107	0.6607
0.0125	21.39	7700	2.2925	0.7096	0.6593
0.0125	21.67	7800	2.3717	0.6998	0.6486
0.0125	21.94	7900	2.4211	0.6951	0.6479
0.0104	22.22	8000	2.3714	0.6978	0.6434
0.0104	22.5	8100	2.3995	0.7004	0.6503
0.0104	22.78	8200	2.3877	0.7044	0.6521
0.0104	23.06	8300	2.4957	0.6972	0.6482
0.0104	23.33	8400	2.2553	0.7180	0.6591
0.0061	23.61	8500	2.3877	0.7068	0.6560
0.0061	23.89	8600	2.4298	0.7036	0.6557
0.0061	24.17	8700	2.3903	0.7055	0.6516
0.0061	24.44	8800	2.3298	0.7065	0.6493
0.0061	24.72	8900	2.3245	0.7110	0.6535
0.0054	25.0	9000	2.3287	0.7086	0.6494
0.0054	25.28	9100	2.4519	0.6989	0.6427
0.0054	25.56	9200	2.4671	0.6988	0.6421
0.0054	25.83	9300	2.5166	0.6955	0.6447
0.0054	26.11	9400	2.4190	0.7056	0.6500
0.0029	26.39	9500	2.4361	0.7049	0.6511
0.0029	26.67	9600	2.4765	0.7029	0.6496
0.0029	26.94	9700	2.5246	0.6988	0.6460
0.0029	27.22	9800	2.4363	0.7051	0.6491
0.0029	27.5	9900	2.4066	0.7075	0.6514
0.0025	27.78	10000	2.3870	0.7092	0.6556
0.0025	28.06	10100	2.4028	0.7081	0.6539
0.0025	28.33	10200	2.3983	0.7080	0.6537
0.0025	28.61	10300	2.3876	0.7088	0.6552
0.0025	28.89	10400	2.4032	0.7080	0.6542
0.0025	29.17	10500	2.4138	0.7081	0.6544
0.0025	29.44	10600	2.3880	0.7098	0.6555
0.0025	29.72	10700	2.3801	0.7100	0.6552
0.0025	30.0	10800	2.3802	0.7101	0.6551

Framework versions

• Transformers 4.33.3
• Pytorch 2.1.1+cu121
• Datasets 2.14.5
• Tokenizers 0.13.3