metadata

language:
  - en
base_model: gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48
tags:
  - generated_from_trainer
datasets:
  - glue
metrics:
  - accuracy
  - f1
model-index:
  - name: hBERTv1_new_pretrain_w_init_48_ver2_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: 0.7181372549019608
          - name: F1
            type: f1
            value: 0.8099173553719009

hBERTv1_new_pretrain_w_init_48_ver2_mrpc

This model is a fine-tuned version of gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48 on the GLUE MRPC dataset. It achieves the following results on the evaluation set:

Loss: 0.5916
Accuracy: 0.7181
F1: 0.8099
Combined Score: 0.7640

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: 64
eval_batch_size: 64
seed: 10
distributed_type: multi-GPU
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Combined Score
0.6666	1.0	58	0.6274	0.6912	0.8006	0.7459
0.6239	2.0	116	0.5916	0.7181	0.8099	0.7640
0.5981	3.0	174	0.6532	0.6225	0.7004	0.6615
0.518	4.0	232	0.6251	0.7108	0.8059	0.7584
0.3848	5.0	290	0.7553	0.6814	0.7869	0.7341
0.2708	6.0	348	1.0696	0.6838	0.7994	0.7416
0.2062	7.0	406	1.2159	0.6103	0.7145	0.6624

Framework versions

Transformers 4.34.0
Pytorch 1.14.0a0+410ce96
Datasets 2.14.5
Tokenizers 0.14.1