metadata

tags:
  - generated_from_trainer
datasets:
  - glue
metrics:
  - accuracy
  - f1
model-index:
  - name: hBERTv2_new_no_pretrain_qqp
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: glue
          type: glue
          config: qqp
          split: validation
          args: qqp
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.6318327974276527
          - name: F1
            type: f1
            value: 0

hBERTv2_new_no_pretrain_qqp

This model is a fine-tuned version of on the glue dataset. It achieves the following results on the evaluation set:

Loss: 0.6580
Accuracy: 0.6318
F1: 0.0
Combined Score: 0.3159

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: 0.0005
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
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Combined Score
0.6669	1.0	2843	0.6595	0.6318	0.3159
0.6591	2.0	5686	0.6587	0.6318	0.3159
0.6589	3.0	8529	0.6582	0.6318	0.3159
0.6587	4.0	11372	0.6580	0.6318	0.3159
0.6586	5.0	14215	0.6579	0.6318	0.3159
0.6586	6.0	17058	0.6580	0.6318	0.3159
0.6586	7.0	19901	0.6580	0.6318	0.3159
0.6586	8.0	22744	0.6579	0.6318	0.3159
0.6586	9.0	25587	0.6580	0.6318	0.3159
0.6586	10.0	28430	0.6580	0.6318	0.3159

Framework versions

Transformers 4.29.2
Pytorch 1.14.0a0+410ce96
Datasets 2.12.0
Tokenizers 0.13.3