Ranjit/final_en

massive_indo

This model is a fine-tuned version of xxxxxxxxx on the massive dataset. It achieves the following results on the evaluation set:

• Loss: 1.0967
• F1: 0.8702

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: 50

Training results

Training Loss	Epoch	Step	Validation Loss	F1
0.747	1.39	500	1.0303	0.5703
0.5618	2.78	1000	0.9201	0.6479
0.3695	4.17	1500	0.8216	0.6990
0.3392	5.56	2000	0.7637	0.7335
0.2638	6.94	2500	0.8244	0.7678
0.1907	8.33	3000	0.7912	0.7979
0.1661	9.72	3500	0.8266	0.7835
0.1073	11.11	4000	0.8120	0.8139
0.1265	12.5	4500	0.8336	0.8344
0.0481	13.89	5000	0.8240	0.8518
0.0646	15.28	5500	0.9290	0.8333
0.0846	16.67	6000	0.9176	0.8461
0.0228	18.06	6500	0.9600	0.8529
0.0696	19.44	7000	0.9769	0.8525
0.0614	20.83	7500	0.9944	0.8545
0.0173	22.22	8000	1.0110	0.8550
0.004	23.61	8500	1.0140	0.8417
0.0032	25.0	9000	1.0771	0.8314
0.0453	26.39	9500	1.0173	0.8424
0.0471	27.78	10000	1.0068	0.8652
0.0128	29.17	10500	1.0595	0.8658
0.0027	30.56	11000	1.0596	0.8506
0.0198	31.94	11500	1.0468	0.8593
0.0027	33.33	12000	1.0537	0.8693
0.0114	34.72	12500	1.0512	0.8620
0.015	36.11	13000	1.0425	0.8813
0.005	37.5	13500	1.1092	0.8749
0.0038	38.89	14000	1.0829	0.8637
0.0096	40.28	14500	1.0902	0.8794
0.0007	41.67	15000	1.0994	0.8651
0.0109	43.06	15500	1.0957	0.8782
0.0026	44.44	16000	1.0997	0.8643
0.0061	45.83	16500	1.0853	0.8672
0.0005	47.22	17000	1.1082	0.8694
0.0005	48.61	17500	1.1016	0.8696
0.0028	50.0	18000	1.0967	0.8702

Framework versions

• Transformers 4.34.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.0