metadata
license: apache-2.0
base_model: google/bert_uncased_L-8_H-256_A-4
tags:
- generated_from_trainer
datasets:
- ag_news
metrics:
- accuracy
model-index:
- name: distillbert-distilled-ag-news-2
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: ag_news
type: ag_news
config: default
split: train
args: default
metrics:
- name: Accuracy
type: accuracy
value: 0.9407916666666667
distillbert-distilled-ag-news-2
This model is a fine-tuned version of google/bert_uncased_L-8_H-256_A-4 on the ag_news dataset. It achieves the following results on the evaluation set:
- Loss: 0.1945
- Accuracy: 0.9408
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: 2e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.238 | 1.0 | 3000 | 0.2240 | 0.9237 |
0.1873 | 2.0 | 6000 | 0.2009 | 0.9329 |
0.1597 | 3.0 | 9000 | 0.1919 | 0.9377 |
0.1495 | 4.0 | 12000 | 0.1948 | 0.9400 |
0.1303 | 5.0 | 15000 | 0.1945 | 0.9408 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1