metadata

language:
  - en
tags:
  - generated_from_trainer
datasets:
  - glue
metrics:
  - accuracy
  - f1
model-index:
  - name: hBERTv1_new_pretrain_w_init__mrpc
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: GLUE MRPC
          type: glue
          config: mrpc
          split: validation
          args: mrpc
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.6862745098039216
          - name: F1
            type: f1
            value: 0.7894736842105262

hBERTv1_new_pretrain_w_init__mrpc

This model is a fine-tuned version of gokuls/bert_12_layer_model_v1_complete_training_new_wt_init on the GLUE MRPC dataset. It achieves the following results on the evaluation set:

Loss: 0.6082
Accuracy: 0.6863
F1: 0.7895
Combined Score: 0.7379

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: 4e-05
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

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Combined Score
0.7111	1.0	29	0.6564	0.6838	0.8122	0.7480
0.6641	2.0	58	0.6160	0.6838	0.8122	0.7480
0.6156	3.0	87	0.6354	0.6838	0.8122	0.7480
0.5817	4.0	116	0.6082	0.6863	0.7895	0.7379
0.5091	5.0	145	0.7812	0.5074	0.5157	0.5115
0.3973	6.0	174	0.7949	0.6544	0.7565	0.7054
0.2966	7.0	203	1.0388	0.6078	0.6887	0.6483
0.2024	8.0	232	1.0065	0.6201	0.7124	0.6663
0.1621	9.0	261	1.3076	0.5735	0.6575	0.6155

Framework versions

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