mobilebert_sa_GLUE_Experiment_data_aug_mrpc

This model is a fine-tuned version of google/mobilebert-uncased on the GLUE MRPC dataset. It achieves the following results on the evaluation set:

• Loss: 0.0000
• Accuracy: 1.0
• F1: 1.0
• Combined Score: 1.0

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: 128
• eval_batch_size: 128
• seed: 10
• distributed_type: multi-GPU
• 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	Accuracy	F1	Combined Score
0.1838	1.0	1959	0.0138	0.9951	0.9964	0.9958
0.0406	2.0	3918	0.0055	1.0	1.0	1.0
0.0267	3.0	5877	0.0129	0.9975	0.9982	0.9979
0.0151	4.0	7836	0.0004	1.0	1.0	1.0
0.0108	5.0	9795	0.0104	0.9975	0.9982	0.9979
0.0075	6.0	11754	0.0000	1.0	1.0	1.0
0.0059	7.0	13713	0.0005	1.0	1.0	1.0
0.0047	8.0	15672	0.0000	1.0	1.0	1.0
0.0033	9.0	17631	0.0001	1.0	1.0	1.0
0.0031	10.0	19590	0.0000	1.0	1.0	1.0
0.0025	11.0	21549	0.0000	1.0	1.0	1.0
0.0019	12.0	23508	0.0000	1.0	1.0	1.0
0.0019	13.0	25467	0.0000	1.0	1.0	1.0
0.0014	14.0	27426	0.0000	1.0	1.0	1.0
0.001	15.0	29385	0.0000	1.0	1.0	1.0
0.001	16.0	31344	0.0000	1.0	1.0	1.0
0.0009	17.0	33303	0.0000	1.0	1.0	1.0
0.0009	18.0	35262	0.0000	1.0	1.0	1.0
0.0006	19.0	37221	0.0000	1.0	1.0	1.0
0.0006	20.0	39180	0.0000	1.0	1.0	1.0
0.0003	21.0	41139	0.0000	1.0	1.0	1.0
0.0003	22.0	43098	0.0000	1.0	1.0	1.0
0.0005	23.0	45057	0.0000	1.0	1.0	1.0

Framework versions

• Transformers 4.26.0
• Pytorch 1.14.0a0+410ce96
• Datasets 2.9.0
• Tokenizers 0.13.2