README.md

metadata

tags:
  - generated_from_trainer
datasets:
  - glue
metrics:
  - accuracy
model-index:
  - name: hBERTv2_new_pretrain_w_init__sst2
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: glue
          type: glue
          config: sst2
          split: validation
          args: sst2
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8245412844036697

hBERTv2_new_pretrain_w_init__sst2

This model is a fine-tuned version of gokuls/bert_12_layer_model_v2_complete_training_new_wt_init on the glue dataset. It achieves the following results on the evaluation set:

• Loss: 0.6070
• Accuracy: 0.8245

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
0.3399	1.0	527	0.4105	0.8360
0.2098	2.0	1054	0.4837	0.8222
0.1578	3.0	1581	0.5173	0.8119
0.1219	4.0	2108	0.5737	0.8337
0.0978	5.0	2635	0.5374	0.8165
0.0803	6.0	3162	0.6070	0.8245

Framework versions

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