hBERTv1_qqp – AI Model by gokuls | AlphaNeural AI

You can deploy this model and start earning money today!

gokuls/
hBERTv1_qqp

like

0

text-classification

generated_from_trainer

autotrain_compatible

endpoints_compatible

Views

No views yet

hBERTv1_qqp

This model is a fine-tuned version of gokuls/bert_12_layer_model_v1 on the GLUE QQP dataset. It achieves the following results on the evaluation set:

Loss: 0.3039
Accuracy: 0.8680
F1: 0.8222
Combined Score: 0.8451

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: 256
eval_batch_size: 256
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
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Combined Score
0.4011	1.0	1422	0.3665	0.8286	0.7947	0.8116
0.3026	2.0	2844	0.3111	0.8625	0.8171	0.8398
0.2472	3.0	4266	0.3039	0.8680	0.8222	0.8451
0.1983	4.0	5688	0.3232	0.8737	0.8327	0.8532
0.157	5.0	7110	0.3742	0.8717	0.8194	0.8456
0.1251	6.0	8532	0.4009	0.8716	0.8146	0.8431
0.1009	7.0	9954	0.4471	0.8699	0.8300	0.8500
0.0828	8.0	11376	0.4176	0.8781	0.8354	0.8568

Framework versions

Transformers 4.26.1
Pytorch 1.14.0a0+410ce96
Datasets 2.10.1
Tokenizers 0.13.2