metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- accuracy
- f1
model-index:
- name: glue-qqp
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: glue
type: glue
args: qqp
metrics:
- name: Accuracy
type: accuracy
value: 0.9033391046252782
- name: F1
type: f1
value: 0.8703384207033842
- task:
type: natural-language-inference
name: Natural Language Inference
dataset:
name: glue
type: glue
config: qqp
split: validation
metrics:
- name: Accuracy
type: accuracy
value: 0.9034627751669553
verified: true
- name: Precision
type: precision
value: 0.8600183582480986
verified: true
- name: Recall
type: recall
value: 0.8812227074235808
verified: true
- name: AUC
type: auc
value: 0.960566250301703
verified: true
- name: F1
type: f1
value: 0.8704914225038989
verified: true
- name: loss
type: loss
value: 0.47982412576675415
verified: true
glue-qqp
This model is a fine-tuned version of distilbert-base-uncased on the glue dataset. It achieves the following results on the evaluation set:
- Loss: 0.4798
- Accuracy: 0.9033
- F1: 0.8703
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: 16
- eval_batch_size: 16
- 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 | F1 |
|---|---|---|---|---|---|
| 0.2779 | 1.0 | 22741 | 0.2697 | 0.8871 | 0.8494 |
| 0.2183 | 2.0 | 45482 | 0.2651 | 0.8966 | 0.8634 |
| 0.1635 | 3.0 | 68223 | 0.3116 | 0.9013 | 0.8685 |
| 0.1312 | 4.0 | 90964 | 0.4102 | 0.9030 | 0.8694 |
| 0.0802 | 5.0 | 113705 | 0.4798 | 0.9033 | 0.8703 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.11.0
- Datasets 2.3.2
- Tokenizers 0.11.6