metadata
library_name: transformers
language:
- en
base_model: Hartunka/tiny_bert_km_20_v2
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- accuracy
model-index:
- name: tiny_bert_km_20_v2_mnli
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: GLUE MNLI
type: glue
args: mnli
metrics:
- name: Accuracy
type: accuracy
value: 0.6295768917819365
tiny_bert_km_20_v2_mnli
This model is a fine-tuned version of Hartunka/tiny_bert_km_20_v2 on the GLUE MNLI dataset. It achieves the following results on the evaluation set:
- Loss: 0.8293
- Accuracy: 0.6296
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: 256
- eval_batch_size: 256
- seed: 10
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.9993 | 1.0 | 1534 | 0.9376 | 0.5527 |
| 0.9067 | 2.0 | 3068 | 0.8824 | 0.5892 |
| 0.8527 | 3.0 | 4602 | 0.8542 | 0.6076 |
| 0.8046 | 4.0 | 6136 | 0.8476 | 0.6214 |
| 0.7574 | 5.0 | 7670 | 0.8323 | 0.6301 |
| 0.7096 | 6.0 | 9204 | 0.8395 | 0.6353 |
| 0.666 | 7.0 | 10738 | 0.8471 | 0.6446 |
| 0.6211 | 8.0 | 12272 | 0.8765 | 0.6424 |
| 0.5794 | 9.0 | 13806 | 0.9187 | 0.6422 |
| 0.539 | 10.0 | 15340 | 0.9313 | 0.6425 |
Framework versions
- Transformers 4.50.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.21.1