metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- cifar10_quality_drift
metrics:
- accuracy
- f1
base_model: microsoft/resnet-50
model-index:
- name: resnet-50-cifar10-quality-drift
results:
- task:
type: image-classification
name: Image Classification
dataset:
name: cifar10_quality_drift
type: cifar10_quality_drift
args: default
metrics:
- type: accuracy
value: 0.724
name: Accuracy
- type: f1
value: 0.7221970011456912
name: F1
resnet-50-cifar10-quality-drift
This model is a fine-tuned version of microsoft/resnet-50 on the cifar10_quality_drift dataset. It achieves the following results on the evaluation set:
- Loss: 0.8235
- Accuracy: 0.724
- F1: 0.7222
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: 0.0002
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- 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 |
|---|---|---|---|---|---|
| 1.7311 | 1.0 | 750 | 1.1310 | 0.6333 | 0.6300 |
| 1.1728 | 2.0 | 1500 | 0.8495 | 0.7153 | 0.7155 |
| 1.0322 | 3.0 | 2250 | 0.8235 | 0.724 | 0.7222 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.12.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1