Model save
Browse files
README.md
ADDED
|
@@ -0,0 +1,79 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
metrics:
|
| 7 |
+
- accuracy
|
| 8 |
+
- f1
|
| 9 |
+
- precision
|
| 10 |
+
- recall
|
| 11 |
+
model-index:
|
| 12 |
+
- name: vit-base-patch16-224-in21k-finetuned-hongrui_mammogram_v_1
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# vit-base-patch16-224-in21k-finetuned-hongrui_mammogram_v_1
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 0.7419
|
| 24 |
+
- Accuracy: 0.6991
|
| 25 |
+
- F1: 0.6767
|
| 26 |
+
- Precision: 0.6830
|
| 27 |
+
- Recall: 0.6991
|
| 28 |
+
|
| 29 |
+
## Model description
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Intended uses & limitations
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training and evaluation data
|
| 38 |
+
|
| 39 |
+
More information needed
|
| 40 |
+
|
| 41 |
+
## Training procedure
|
| 42 |
+
|
| 43 |
+
### Training hyperparameters
|
| 44 |
+
|
| 45 |
+
The following hyperparameters were used during training:
|
| 46 |
+
- learning_rate: 5e-05
|
| 47 |
+
- train_batch_size: 64
|
| 48 |
+
- eval_batch_size: 64
|
| 49 |
+
- seed: 42
|
| 50 |
+
- gradient_accumulation_steps: 4
|
| 51 |
+
- total_train_batch_size: 256
|
| 52 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 53 |
+
- lr_scheduler_type: linear
|
| 54 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 55 |
+
- num_epochs: 10
|
| 56 |
+
- mixed_precision_training: Native AMP
|
| 57 |
+
|
| 58 |
+
### Training results
|
| 59 |
+
|
| 60 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|
| 61 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
|
| 62 |
+
| 0.8576 | 1.0 | 171 | 0.8431 | 0.6678 | 0.6067 | 0.7751 | 0.6678 |
|
| 63 |
+
| 0.8297 | 2.0 | 342 | 0.7965 | 0.6791 | 0.6182 | 0.6758 | 0.6791 |
|
| 64 |
+
| 0.8303 | 3.0 | 513 | 0.7872 | 0.6842 | 0.6360 | 0.6704 | 0.6842 |
|
| 65 |
+
| 0.7814 | 4.0 | 684 | 0.7717 | 0.6843 | 0.6597 | 0.6601 | 0.6843 |
|
| 66 |
+
| 0.7768 | 5.0 | 855 | 0.7694 | 0.6906 | 0.6544 | 0.6775 | 0.6906 |
|
| 67 |
+
| 0.7415 | 6.0 | 1026 | 0.7572 | 0.6962 | 0.6718 | 0.6764 | 0.6962 |
|
| 68 |
+
| 0.7351 | 7.0 | 1197 | 0.7549 | 0.6922 | 0.6569 | 0.6648 | 0.6922 |
|
| 69 |
+
| 0.7197 | 8.0 | 1368 | 0.7479 | 0.6986 | 0.6855 | 0.6926 | 0.6986 |
|
| 70 |
+
| 0.7087 | 9.0 | 1539 | 0.7445 | 0.6979 | 0.6697 | 0.6792 | 0.6979 |
|
| 71 |
+
| 0.6977 | 10.0 | 1710 | 0.7419 | 0.6991 | 0.6767 | 0.6830 | 0.6991 |
|
| 72 |
+
|
| 73 |
+
|
| 74 |
+
### Framework versions
|
| 75 |
+
|
| 76 |
+
- Transformers 4.42.4
|
| 77 |
+
- Pytorch 2.4.0+cu121
|
| 78 |
+
- Datasets 2.21.0
|
| 79 |
+
- Tokenizers 0.19.1
|