metadata
base_model: gokuls/HBERTv1_48_L12_H512_A8
tags:
- generated_from_trainer
datasets:
- emotion
metrics:
- accuracy
model-index:
- name: HBERTv1_48_L12_H512_A8_emotion
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: 0.8965
HBERTv1_48_L12_H512_A8_emotion
This model is a fine-tuned version of gokuls/HBERTv1_48_L12_H512_A8 on the emotion dataset. It achieves the following results on the evaluation set:
- Loss: 0.3811
- Accuracy: 0.8965
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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.8277 | 1.0 | 250 | 0.4343 | 0.852 |
| 0.3876 | 2.0 | 500 | 0.3535 | 0.877 |
| 0.297 | 3.0 | 750 | 0.2943 | 0.887 |
| 0.229 | 4.0 | 1000 | 0.2967 | 0.891 |
| 0.1794 | 5.0 | 1250 | 0.2958 | 0.891 |
| 0.132 | 6.0 | 1500 | 0.3239 | 0.8915 |
| 0.1018 | 7.0 | 1750 | 0.3811 | 0.8965 |
| 0.0711 | 8.0 | 2000 | 0.4557 | 0.892 |
| 0.0504 | 9.0 | 2250 | 0.5029 | 0.894 |
| 0.034 | 10.0 | 2500 | 0.5607 | 0.8935 |
Framework versions
- Transformers 4.34.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.14.5
- Tokenizers 0.14.0