test-bert – AI Model by hung200504 | AlphaNeural AI

You can deploy this model and start earning money today!

hung200504/
test-bert

like

0

question-answering

generated_from_trainer

google-bert/bert-base-cased

endpoints_compatible

Views

No views yet

test-bert

This model is a fine-tuned version of bert-base-cased on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 6.0267

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: 3e-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: 3

Training results

Training Loss	Epoch	Step	Validation Loss
5.8219	0.09	5	5.9435
4.9256	0.18	10	6.1109
4.3213	0.27	15	5.7204
3.5947	0.36	20	5.7525
3.0974	0.45	25	5.6447
2.7481	0.55	30	5.2776
2.207	0.64	35	5.3963
1.8922	0.73	40	5.4622
1.7034	0.82	45	5.3710
1.4313	0.91	50	5.3449
1.0748	1.0	55	5.3580
1.0215	1.09	60	5.4713
0.7634	1.18	65	5.5980
0.7535	1.27	70	5.6049
0.6063	1.36	75	5.5830
0.4824	1.45	80	5.6753
0.48	1.55	85	5.7216
0.4884	1.64	90	5.7817
0.5813	1.73	95	5.9427
0.4287	1.82	100	6.0148
0.4061	1.91	105	5.8752
0.55	2.0	110	5.8723
0.4631	2.09	115	5.8288
0.2987	2.18	120	5.8808
0.3359	2.27	125	5.8884
0.3002	2.36	130	5.9442
0.299	2.45	135	5.9331
0.3393	2.55	140	5.9454
0.2656	2.64	145	5.9894
0.3582	2.73	150	6.0142
0.2111	2.82	155	6.0329
0.2574	2.91	160	6.0304
0.2471	3.0	165	6.0267

Framework versions

Transformers 4.34.1
Pytorch 2.1.0+cu118
Datasets 2.14.5
Tokenizers 0.14.1