metadata
license: mit
base_model: microsoft/MiniLM-L12-H384-uncased
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: minilm-imdb
results: []
minilm-imdb
This model is a fine-tuned version of microsoft/MiniLM-L12-H384-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2403
- Accuracy: 0.9229
- F1: 0.9228
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: 4
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.1511 | 1.0 | 293 | 0.2212 | 0.9234 | 0.9229 |
| 0.1047 | 2.0 | 586 | 0.2211 | 0.9230 | 0.9217 |
| 0.1008 | 3.0 | 879 | 0.2403 | 0.9229 | 0.9228 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.0
- Tokenizers 0.15.0