Commit
•
ec95a74
1
Parent(s):
a9122c2
Model save
Browse files
README.md
ADDED
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
library_name: peft
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
datasets:
|
7 |
+
- medmnist-v2
|
8 |
+
metrics:
|
9 |
+
- accuracy
|
10 |
+
- precision
|
11 |
+
- recall
|
12 |
+
- f1
|
13 |
+
base_model: facebook/deit-base-patch16-224
|
14 |
+
model-index:
|
15 |
+
- name: organc-deit-base-finetuned
|
16 |
+
results: []
|
17 |
+
---
|
18 |
+
|
19 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
20 |
+
should probably proofread and complete it, then remove this comment. -->
|
21 |
+
|
22 |
+
# organc-deit-base-finetuned
|
23 |
+
|
24 |
+
This model is a fine-tuned version of [facebook/deit-base-patch16-224](https://huggingface.co/facebook/deit-base-patch16-224) on the medmnist-v2 dataset.
|
25 |
+
It achieves the following results on the evaluation set:
|
26 |
+
- Loss: 0.0745
|
27 |
+
- Accuracy: 0.9870
|
28 |
+
- Precision: 0.9877
|
29 |
+
- Recall: 0.9863
|
30 |
+
- F1: 0.9868
|
31 |
+
|
32 |
+
## Model description
|
33 |
+
|
34 |
+
More information needed
|
35 |
+
|
36 |
+
## Intended uses & limitations
|
37 |
+
|
38 |
+
More information needed
|
39 |
+
|
40 |
+
## Training and evaluation data
|
41 |
+
|
42 |
+
More information needed
|
43 |
+
|
44 |
+
## Training procedure
|
45 |
+
|
46 |
+
### Training hyperparameters
|
47 |
+
|
48 |
+
The following hyperparameters were used during training:
|
49 |
+
- learning_rate: 0.005
|
50 |
+
- train_batch_size: 16
|
51 |
+
- eval_batch_size: 16
|
52 |
+
- seed: 42
|
53 |
+
- gradient_accumulation_steps: 4
|
54 |
+
- total_train_batch_size: 64
|
55 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
56 |
+
- lr_scheduler_type: linear
|
57 |
+
- num_epochs: 10
|
58 |
+
- mixed_precision_training: Native AMP
|
59 |
+
|
60 |
+
### Training results
|
61 |
+
|
62 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|
63 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
|
64 |
+
| 0.7947 | 1.0 | 203 | 0.3123 | 0.8976 | 0.9090 | 0.8450 | 0.8632 |
|
65 |
+
| 0.6703 | 2.0 | 406 | 0.1400 | 0.9607 | 0.9590 | 0.9543 | 0.9535 |
|
66 |
+
| 0.5941 | 3.0 | 609 | 0.1182 | 0.9699 | 0.9647 | 0.9681 | 0.9649 |
|
67 |
+
| 0.5837 | 4.0 | 813 | 0.1016 | 0.9678 | 0.9558 | 0.9586 | 0.9551 |
|
68 |
+
| 0.5193 | 5.0 | 1016 | 0.0800 | 0.9791 | 0.9701 | 0.9684 | 0.9675 |
|
69 |
+
| 0.5513 | 6.0 | 1219 | 0.0579 | 0.9862 | 0.9831 | 0.9855 | 0.9840 |
|
70 |
+
| 0.4343 | 7.0 | 1422 | 0.0775 | 0.9833 | 0.9858 | 0.9818 | 0.9835 |
|
71 |
+
| 0.3942 | 8.0 | 1626 | 0.0782 | 0.9833 | 0.9813 | 0.9827 | 0.9817 |
|
72 |
+
| 0.2971 | 9.0 | 1829 | 0.0839 | 0.9862 | 0.9884 | 0.9866 | 0.9873 |
|
73 |
+
| 0.3242 | 9.99 | 2030 | 0.0745 | 0.9870 | 0.9877 | 0.9863 | 0.9868 |
|
74 |
+
|
75 |
+
|
76 |
+
### Framework versions
|
77 |
+
|
78 |
+
- PEFT 0.10.0
|
79 |
+
- Transformers 4.38.2
|
80 |
+
- Pytorch 2.2.1+cu121
|
81 |
+
- Datasets 2.18.0
|
82 |
+
- Tokenizers 0.15.2
|