metadata
license: apache-2.0
library_name: peft
tags:
- generated_from_trainer
datasets:
- medmnist-v2
metrics:
- accuracy
- precision
- recall
- f1
base_model: microsoft/beit-base-patch16-224-pt22k-ft22k
model-index:
- name: derma-beit-base-finetuned
results: []
derma-beit-base-finetuned
This model is a fine-tuned version of microsoft/beit-base-patch16-224-pt22k-ft22k on the medmnist-v2 dataset. It achieves the following results on the evaluation set:
- Loss: 0.6096
- Accuracy: 0.7727
- Precision: 0.6427
- Recall: 0.5346
- F1: 0.5283
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: 0.005
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
---|---|---|---|---|---|---|---|
0.9135 | 1.0 | 109 | 0.7698 | 0.7198 | 0.5179 | 0.3103 | 0.3050 |
0.8352 | 2.0 | 219 | 0.7352 | 0.7298 | 0.5362 | 0.4231 | 0.3884 |
0.7891 | 3.0 | 328 | 0.7575 | 0.7178 | 0.3954 | 0.4000 | 0.3667 |
0.7649 | 4.0 | 438 | 0.6879 | 0.7418 | 0.5009 | 0.3972 | 0.4146 |
0.8146 | 5.0 | 547 | 0.7471 | 0.7178 | 0.4490 | 0.4141 | 0.3641 |
0.6831 | 6.0 | 657 | 0.7007 | 0.7368 | 0.4777 | 0.4148 | 0.4252 |
0.695 | 7.0 | 766 | 0.6797 | 0.7428 | 0.4638 | 0.5334 | 0.4841 |
0.6646 | 8.0 | 876 | 0.6534 | 0.7537 | 0.6130 | 0.5077 | 0.4933 |
0.675 | 9.0 | 985 | 0.6238 | 0.7667 | 0.6518 | 0.5431 | 0.5308 |
0.6145 | 9.95 | 1090 | 0.6096 | 0.7727 | 0.6427 | 0.5346 | 0.5283 |
Framework versions
- PEFT 0.9.0
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2