README.md

metadata

base_model: gokuls/HBERTv1_48_L10_H768_A12
tags:
  - generated_from_trainer
datasets:
  - emotion
metrics:
  - accuracy
model-index:
  - name: hbertv1-emotion-intermediate_KD_new_2
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: emotion
          type: emotion
          config: split
          split: validation
          args: split
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.897

hbertv1-emotion-intermediate_KD_new_2

This model is a fine-tuned version of gokuls/HBERTv1_48_L10_H768_A12 on the emotion dataset. It achieves the following results on the evaluation set:

• Loss: 1.2346
• Accuracy: 0.897

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: 5e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 33
• 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
2.954	1.0	250	1.8290	0.8255
1.775	2.0	500	1.6132	0.8285
1.5532	3.0	750	1.4540	0.8515
1.4101	4.0	1000	1.3212	0.8855
1.3138	5.0	1250	1.2489	0.8935
1.2434	6.0	1500	1.2280	0.896
1.1933	7.0	1750	1.2346	0.897
1.1417	8.0	2000	1.2159	0.8835
1.0954	9.0	2250	1.2792	0.8855
1.056	10.0	2500	1.2294	0.8875
1.0235	11.0	2750	1.2474	0.883
0.9943	12.0	3000	1.2179	0.886

Framework versions

• Transformers 4.35.2
• Pytorch 1.14.0a0+410ce96
• Datasets 2.15.0
• Tokenizers 0.15.0