|
---
|
|
library_name: transformers
|
|
license: apache-2.0
|
|
base_model: google/vit-base-patch16-224-in21k
|
|
tags:
|
|
- generated_from_trainer
|
|
datasets:
|
|
- imagefolder
|
|
metrics:
|
|
- accuracy
|
|
- precision
|
|
- recall
|
|
- f1
|
|
model-index:
|
|
- name: vit-base-kidney-stone-Michel_Daudon_-w256_1k_v1-_MIX
|
|
results:
|
|
- task:
|
|
name: Image Classification
|
|
type: image-classification
|
|
dataset:
|
|
name: imagefolder
|
|
type: imagefolder
|
|
config: default
|
|
split: test
|
|
args: default
|
|
metrics:
|
|
- name: Accuracy
|
|
type: accuracy
|
|
value: 0.83375
|
|
- name: Precision
|
|
type: precision
|
|
value: 0.8588680878951838
|
|
- name: Recall
|
|
type: recall
|
|
value: 0.83375
|
|
- name: F1
|
|
type: f1
|
|
value: 0.8355968544321966
|
|
---
|
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
should probably proofread and complete it, then remove this comment. -->
|
|
|
|
# vit-base-kidney-stone-Michel_Daudon_-w256_1k_v1-_MIX
|
|
|
|
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset.
|
|
It achieves the following results on the evaluation set:
|
|
- Loss: 0.4940
|
|
- Accuracy: 0.8337
|
|
- Precision: 0.8589
|
|
- Recall: 0.8337
|
|
- F1: 0.8356
|
|
|
|
## 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: 32
|
|
- eval_batch_size: 8
|
|
- seed: 42
|
|
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
|
- lr_scheduler_type: linear
|
|
- num_epochs: 15
|
|
- mixed_precision_training: Native AMP
|
|
|
|
### Training results
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|
|
|:-------------:|:-------:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
|
|
| 0.1919 | 0.3333 | 100 | 0.4940 | 0.8337 | 0.8589 | 0.8337 | 0.8356 |
|
|
| 0.1697 | 0.6667 | 200 | 0.6993 | 0.8092 | 0.8485 | 0.8092 | 0.8059 |
|
|
| 0.1514 | 1.0 | 300 | 0.5555 | 0.8442 | 0.8565 | 0.8442 | 0.8443 |
|
|
| 0.0991 | 1.3333 | 400 | 0.5918 | 0.8467 | 0.8741 | 0.8467 | 0.8453 |
|
|
| 0.0415 | 1.6667 | 500 | 0.6080 | 0.8558 | 0.8690 | 0.8558 | 0.8553 |
|
|
| 0.1112 | 2.0 | 600 | 0.9788 | 0.7983 | 0.8485 | 0.7983 | 0.8028 |
|
|
| 0.0658 | 2.3333 | 700 | 1.0272 | 0.8004 | 0.8310 | 0.8004 | 0.8002 |
|
|
| 0.0977 | 2.6667 | 800 | 0.6861 | 0.8479 | 0.8570 | 0.8479 | 0.8482 |
|
|
| 0.03 | 3.0 | 900 | 0.8317 | 0.8025 | 0.8225 | 0.8025 | 0.8048 |
|
|
| 0.0253 | 3.3333 | 1000 | 0.8574 | 0.8242 | 0.8408 | 0.8242 | 0.8254 |
|
|
| 0.0564 | 3.6667 | 1100 | 0.8591 | 0.8392 | 0.8513 | 0.8392 | 0.8343 |
|
|
| 0.0285 | 4.0 | 1200 | 1.3453 | 0.7512 | 0.8090 | 0.7512 | 0.7484 |
|
|
| 0.002 | 4.3333 | 1300 | 0.9746 | 0.8192 | 0.8381 | 0.8192 | 0.8227 |
|
|
| 0.0214 | 4.6667 | 1400 | 0.7404 | 0.8646 | 0.8641 | 0.8646 | 0.8572 |
|
|
| 0.0282 | 5.0 | 1500 | 1.0063 | 0.8233 | 0.8486 | 0.8233 | 0.8219 |
|
|
| 0.03 | 5.3333 | 1600 | 1.0066 | 0.8025 | 0.8376 | 0.8025 | 0.8058 |
|
|
| 0.028 | 5.6667 | 1700 | 1.1451 | 0.8108 | 0.8325 | 0.8108 | 0.8067 |
|
|
| 0.0078 | 6.0 | 1800 | 1.0700 | 0.805 | 0.8220 | 0.805 | 0.8045 |
|
|
| 0.0008 | 6.3333 | 1900 | 1.0180 | 0.8146 | 0.8303 | 0.8146 | 0.8165 |
|
|
| 0.0008 | 6.6667 | 2000 | 0.9882 | 0.8246 | 0.8401 | 0.8246 | 0.8236 |
|
|
| 0.0006 | 7.0 | 2100 | 1.0366 | 0.8283 | 0.8424 | 0.8283 | 0.8270 |
|
|
| 0.0009 | 7.3333 | 2200 | 1.1136 | 0.8121 | 0.8309 | 0.8121 | 0.8143 |
|
|
| 0.0068 | 7.6667 | 2300 | 1.0873 | 0.8117 | 0.8128 | 0.8117 | 0.8015 |
|
|
| 0.0006 | 8.0 | 2400 | 0.8601 | 0.8325 | 0.8383 | 0.8325 | 0.8292 |
|
|
| 0.0187 | 8.3333 | 2500 | 0.9700 | 0.8258 | 0.8375 | 0.8258 | 0.8241 |
|
|
| 0.0005 | 8.6667 | 2600 | 0.8825 | 0.8175 | 0.8339 | 0.8175 | 0.8199 |
|
|
| 0.0005 | 9.0 | 2700 | 1.0314 | 0.8242 | 0.8455 | 0.8242 | 0.8230 |
|
|
| 0.0004 | 9.3333 | 2800 | 1.0323 | 0.8233 | 0.8443 | 0.8233 | 0.8230 |
|
|
| 0.0003 | 9.6667 | 2900 | 1.0397 | 0.8229 | 0.8433 | 0.8229 | 0.8229 |
|
|
| 0.0003 | 10.0 | 3000 | 1.0473 | 0.8237 | 0.8437 | 0.8237 | 0.8239 |
|
|
| 0.0003 | 10.3333 | 3100 | 1.0536 | 0.8229 | 0.8428 | 0.8229 | 0.8233 |
|
|
| 0.0003 | 10.6667 | 3200 | 1.0605 | 0.8229 | 0.8429 | 0.8229 | 0.8234 |
|
|
| 0.0003 | 11.0 | 3300 | 1.0667 | 0.8229 | 0.8429 | 0.8229 | 0.8234 |
|
|
| 0.0002 | 11.3333 | 3400 | 1.0711 | 0.8237 | 0.8436 | 0.8237 | 0.8243 |
|
|
| 0.0002 | 11.6667 | 3500 | 1.0750 | 0.8246 | 0.8441 | 0.8246 | 0.8251 |
|
|
| 0.0002 | 12.0 | 3600 | 1.0804 | 0.825 | 0.8443 | 0.825 | 0.8257 |
|
|
| 0.0002 | 12.3333 | 3700 | 1.0839 | 0.825 | 0.8440 | 0.825 | 0.8257 |
|
|
| 0.0002 | 12.6667 | 3800 | 1.0875 | 0.8246 | 0.8436 | 0.8246 | 0.8253 |
|
|
| 0.0002 | 13.0 | 3900 | 1.0909 | 0.8246 | 0.8436 | 0.8246 | 0.8253 |
|
|
| 0.0002 | 13.3333 | 4000 | 1.0930 | 0.8246 | 0.8436 | 0.8246 | 0.8253 |
|
|
| 0.0002 | 13.6667 | 4100 | 1.0954 | 0.8237 | 0.8429 | 0.8237 | 0.8246 |
|
|
| 0.0002 | 14.0 | 4200 | 1.0975 | 0.8237 | 0.8429 | 0.8237 | 0.8246 |
|
|
| 0.0002 | 14.3333 | 4300 | 1.0988 | 0.8237 | 0.8429 | 0.8237 | 0.8246 |
|
|
| 0.0002 | 14.6667 | 4400 | 1.0997 | 0.8237 | 0.8429 | 0.8237 | 0.8246 |
|
|
| 0.0002 | 15.0 | 4500 | 1.1000 | 0.8237 | 0.8429 | 0.8237 | 0.8246 |
|
|
|
|
|
|
### Framework versions
|
|
|
|
- Transformers 4.48.2
|
|
- Pytorch 2.6.0+cu126
|
|
- Datasets 3.2.0
|
|
- Tokenizers 0.21.0
|
|
|
