metadata

base_model: huawei-noah/TinyBERT_General_4L_312D
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: NLP_Capstone
    results: []

NLP_Capstone

This model is a fine-tuned version of huawei-noah/TinyBERT_General_4L_312D on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.3176
Accuracy: 0.8671

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: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5286	0.2	500	0.4169	0.8251
0.4299	0.4	1000	0.4137	0.8332
0.3856	0.6	1500	0.3714	0.8512
0.3692	0.8	2000	0.3176	0.8671
0.3604	1.0	2500	0.3869	0.8635
0.3457	1.2	3000	0.4126	0.8631
0.3291	1.41	3500	0.4272	0.8675
0.3481	1.61	4000	0.3754	0.8775
0.3253	1.81	4500	0.4293	0.8649
0.3306	2.01	5000	0.3807	0.8789
0.2849	2.21	5500	0.4291	0.8803
0.2824	2.41	6000	0.4058	0.8797
0.279	2.61	6500	0.4521	0.8761
0.2944	2.81	7000	0.4986	0.8747
0.3347	3.01	7500	0.4364	0.8815
0.2622	3.21	8000	0.5368	0.8703
0.2494	3.41	8500	0.4795	0.8854
0.2645	3.61	9000	0.4795	0.8864
0.243	3.81	9500	0.4570	0.8874
0.2399	4.01	10000	0.5219	0.8795
0.2103	4.22	10500	0.5325	0.8775
0.2196	4.42	11000	0.5629	0.8729
0.2494	4.62	11500	0.5087	0.8826
0.1968	4.82	12000	0.5332	0.8779

Framework versions

Transformers 4.34.1
Pytorch 2.1.0+cu118
Datasets 2.14.6
Tokenizers 0.14.1