tiny_bert_29_medicare_intents

This model is a fine-tuned version of prajjwal1/bert-tiny on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3859
• Accuracy: 0.9245

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: 16
• eval_batch_size: 16
• 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	Accuracy
No log	1.0	276	3.1352	0.2551
3.2002	2.0	552	2.9038	0.3837
3.2002	3.0	828	2.6758	0.5
2.8069	4.0	1104	2.4689	0.5510
2.8069	5.0	1380	2.2695	0.5796
2.4336	6.0	1656	2.0859	0.6265
2.4336	7.0	1932	1.9171	0.6449
2.1115	8.0	2208	1.7675	0.6592
2.1115	9.0	2484	1.6326	0.6898
1.8187	10.0	2760	1.5044	0.7224
1.5769	11.0	3036	1.3916	0.7469
1.5769	12.0	3312	1.2908	0.7694
1.3698	13.0	3588	1.2009	0.7898
1.3698	14.0	3864	1.1191	0.8122
1.1965	15.0	4140	1.0441	0.8224
1.1965	16.0	4416	0.9788	0.8367
1.0564	17.0	4692	0.9202	0.8510
1.0564	18.0	4968	0.8699	0.8612
0.9545	19.0	5244	0.8236	0.8653
0.8449	20.0	5520	0.7833	0.8673
0.8449	21.0	5796	0.7402	0.8837
0.7507	22.0	6072	0.7087	0.8816
0.7507	23.0	6348	0.6743	0.8837
0.6936	24.0	6624	0.6459	0.8980
0.6936	25.0	6900	0.6198	0.8980
0.6268	26.0	7176	0.5965	0.9
0.6268	27.0	7452	0.5739	0.9061
0.5721	28.0	7728	0.5510	0.9122
0.5298	29.0	8004	0.5345	0.9122
0.5298	30.0	8280	0.5183	0.9143
0.4958	31.0	8556	0.5027	0.9163
0.4958	32.0	8832	0.4892	0.9163
0.4595	33.0	9108	0.4766	0.9163
0.4595	34.0	9384	0.4649	0.9163
0.4268	35.0	9660	0.4527	0.9163
0.4268	36.0	9936	0.4444	0.9184
0.4085	37.0	10212	0.4353	0.9163
0.4085	38.0	10488	0.4271	0.9163
0.3855	39.0	10764	0.4208	0.9163
0.3684	40.0	11040	0.4152	0.9163
0.3684	41.0	11316	0.4098	0.9184
0.3591	42.0	11592	0.4035	0.9184
0.3591	43.0	11868	0.3993	0.9204
0.3416	44.0	12144	0.3950	0.9245
0.3416	45.0	12420	0.3919	0.9224
0.3377	46.0	12696	0.3897	0.9245
0.3377	47.0	12972	0.3887	0.9245
0.3364	48.0	13248	0.3871	0.9245
0.3282	49.0	13524	0.3862	0.9245
0.3282	50.0	13800	0.3859	0.9245

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.1