phuong-tk-nguyen
commited on
Commit
·
0f50716
1
Parent(s):
fe0c68e
Model save
Browse files- README.md +112 -0
- config.json +77 -0
- model.safetensors +3 -0
- preprocessor_config.json +22 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,112 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: microsoft/swin-base-patch4-window7-224-in22k
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
datasets:
|
7 |
+
- imagefolder
|
8 |
+
metrics:
|
9 |
+
- accuracy
|
10 |
+
model-index:
|
11 |
+
- name: swin-base-patch4-window7-224-in22k-finetuned-cifar10
|
12 |
+
results:
|
13 |
+
- task:
|
14 |
+
name: Image Classification
|
15 |
+
type: image-classification
|
16 |
+
dataset:
|
17 |
+
name: imagefolder
|
18 |
+
type: imagefolder
|
19 |
+
config: default
|
20 |
+
split: train
|
21 |
+
args: default
|
22 |
+
metrics:
|
23 |
+
- name: Accuracy
|
24 |
+
type: accuracy
|
25 |
+
value: 0.9858
|
26 |
+
---
|
27 |
+
|
28 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
29 |
+
should probably proofread and complete it, then remove this comment. -->
|
30 |
+
|
31 |
+
# swin-base-patch4-window7-224-in22k-finetuned-cifar10
|
32 |
+
|
33 |
+
This model is a fine-tuned version of [microsoft/swin-base-patch4-window7-224-in22k](https://huggingface.co/microsoft/swin-base-patch4-window7-224-in22k) on the imagefolder dataset.
|
34 |
+
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 0.0414
|
36 |
+
- Accuracy: 0.9858
|
37 |
+
|
38 |
+
## Model description
|
39 |
+
|
40 |
+
More information needed
|
41 |
+
|
42 |
+
## Intended uses & limitations
|
43 |
+
|
44 |
+
More information needed
|
45 |
+
|
46 |
+
## Training and evaluation data
|
47 |
+
|
48 |
+
More information needed
|
49 |
+
|
50 |
+
## Training procedure
|
51 |
+
|
52 |
+
### Training hyperparameters
|
53 |
+
|
54 |
+
The following hyperparameters were used during training:
|
55 |
+
- learning_rate: 5e-05
|
56 |
+
- train_batch_size: 32
|
57 |
+
- eval_batch_size: 32
|
58 |
+
- seed: 42
|
59 |
+
- gradient_accumulation_steps: 4
|
60 |
+
- total_train_batch_size: 128
|
61 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
62 |
+
- lr_scheduler_type: linear
|
63 |
+
- lr_scheduler_warmup_ratio: 0.1
|
64 |
+
- num_epochs: 1
|
65 |
+
|
66 |
+
### Training results
|
67 |
+
|
68 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
69 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
70 |
+
| 2.303 | 0.03 | 10 | 2.1672 | 0.2334 |
|
71 |
+
| 2.0158 | 0.06 | 20 | 1.6672 | 0.657 |
|
72 |
+
| 1.4855 | 0.09 | 30 | 0.8292 | 0.8704 |
|
73 |
+
| 0.7451 | 0.11 | 40 | 0.2578 | 0.93 |
|
74 |
+
| 0.5618 | 0.14 | 50 | 0.1476 | 0.962 |
|
75 |
+
| 0.4545 | 0.17 | 60 | 0.1248 | 0.9642 |
|
76 |
+
| 0.4587 | 0.2 | 70 | 0.0941 | 0.9748 |
|
77 |
+
| 0.3911 | 0.23 | 80 | 0.0944 | 0.9712 |
|
78 |
+
| 0.3839 | 0.26 | 90 | 0.0848 | 0.9756 |
|
79 |
+
| 0.3864 | 0.28 | 100 | 0.0744 | 0.978 |
|
80 |
+
| 0.3141 | 0.31 | 110 | 0.0673 | 0.98 |
|
81 |
+
| 0.3764 | 0.34 | 120 | 0.0706 | 0.9764 |
|
82 |
+
| 0.3003 | 0.37 | 130 | 0.0600 | 0.984 |
|
83 |
+
| 0.3566 | 0.4 | 140 | 0.0562 | 0.9826 |
|
84 |
+
| 0.2855 | 0.43 | 150 | 0.0567 | 0.9816 |
|
85 |
+
| 0.3351 | 0.45 | 160 | 0.0543 | 0.9828 |
|
86 |
+
| 0.2977 | 0.48 | 170 | 0.0568 | 0.9798 |
|
87 |
+
| 0.2924 | 0.51 | 180 | 0.0577 | 0.9804 |
|
88 |
+
| 0.2884 | 0.54 | 190 | 0.0551 | 0.983 |
|
89 |
+
| 0.3067 | 0.57 | 200 | 0.0487 | 0.983 |
|
90 |
+
| 0.3159 | 0.6 | 210 | 0.0513 | 0.984 |
|
91 |
+
| 0.2795 | 0.63 | 220 | 0.0460 | 0.9846 |
|
92 |
+
| 0.3113 | 0.65 | 230 | 0.0495 | 0.9832 |
|
93 |
+
| 0.2882 | 0.68 | 240 | 0.0475 | 0.9838 |
|
94 |
+
| 0.263 | 0.71 | 250 | 0.0449 | 0.9854 |
|
95 |
+
| 0.2686 | 0.74 | 260 | 0.0510 | 0.9826 |
|
96 |
+
| 0.2705 | 0.77 | 270 | 0.0483 | 0.9846 |
|
97 |
+
| 0.2807 | 0.8 | 280 | 0.0430 | 0.9854 |
|
98 |
+
| 0.2583 | 0.82 | 290 | 0.0452 | 0.9858 |
|
99 |
+
| 0.2346 | 0.85 | 300 | 0.0435 | 0.9858 |
|
100 |
+
| 0.2294 | 0.88 | 310 | 0.0434 | 0.986 |
|
101 |
+
| 0.2608 | 0.91 | 320 | 0.0433 | 0.986 |
|
102 |
+
| 0.2642 | 0.94 | 330 | 0.0425 | 0.9866 |
|
103 |
+
| 0.2781 | 0.97 | 340 | 0.0417 | 0.986 |
|
104 |
+
| 0.247 | 1.0 | 350 | 0.0414 | 0.9858 |
|
105 |
+
|
106 |
+
|
107 |
+
### Framework versions
|
108 |
+
|
109 |
+
- Transformers 4.35.0
|
110 |
+
- Pytorch 2.1.1
|
111 |
+
- Datasets 2.14.6
|
112 |
+
- Tokenizers 0.14.1
|
config.json
ADDED
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "microsoft/swin-base-patch4-window7-224-in22k",
|
3 |
+
"architectures": [
|
4 |
+
"SwinForImageClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.0,
|
7 |
+
"depths": [
|
8 |
+
2,
|
9 |
+
2,
|
10 |
+
18,
|
11 |
+
2
|
12 |
+
],
|
13 |
+
"drop_path_rate": 0.1,
|
14 |
+
"embed_dim": 128,
|
15 |
+
"encoder_stride": 32,
|
16 |
+
"hidden_act": "gelu",
|
17 |
+
"hidden_dropout_prob": 0.0,
|
18 |
+
"hidden_size": 1024,
|
19 |
+
"id2label": {
|
20 |
+
"0": "Airplane",
|
21 |
+
"1": "Automobile",
|
22 |
+
"2": "Bird",
|
23 |
+
"3": "Cat",
|
24 |
+
"4": "Deer",
|
25 |
+
"5": "Dog",
|
26 |
+
"6": "Frog",
|
27 |
+
"7": "Horse",
|
28 |
+
"8": "Ship",
|
29 |
+
"9": "Truck"
|
30 |
+
},
|
31 |
+
"image_size": 224,
|
32 |
+
"initializer_range": 0.02,
|
33 |
+
"label2id": {
|
34 |
+
"Airplane": 0,
|
35 |
+
"Automobile": 1,
|
36 |
+
"Bird": 2,
|
37 |
+
"Cat": 3,
|
38 |
+
"Deer": 4,
|
39 |
+
"Dog": 5,
|
40 |
+
"Frog": 6,
|
41 |
+
"Horse": 7,
|
42 |
+
"Ship": 8,
|
43 |
+
"Truck": 9
|
44 |
+
},
|
45 |
+
"layer_norm_eps": 1e-05,
|
46 |
+
"mlp_ratio": 4.0,
|
47 |
+
"model_type": "swin",
|
48 |
+
"num_channels": 3,
|
49 |
+
"num_heads": [
|
50 |
+
4,
|
51 |
+
8,
|
52 |
+
16,
|
53 |
+
32
|
54 |
+
],
|
55 |
+
"num_layers": 4,
|
56 |
+
"out_features": [
|
57 |
+
"stage4"
|
58 |
+
],
|
59 |
+
"out_indices": [
|
60 |
+
4
|
61 |
+
],
|
62 |
+
"patch_size": 4,
|
63 |
+
"path_norm": true,
|
64 |
+
"problem_type": "single_label_classification",
|
65 |
+
"qkv_bias": true,
|
66 |
+
"stage_names": [
|
67 |
+
"stem",
|
68 |
+
"stage1",
|
69 |
+
"stage2",
|
70 |
+
"stage3",
|
71 |
+
"stage4"
|
72 |
+
],
|
73 |
+
"torch_dtype": "float32",
|
74 |
+
"transformers_version": "4.35.0",
|
75 |
+
"use_absolute_embeddings": false,
|
76 |
+
"window_size": 7
|
77 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0b121fdc3383bdaf42fab0046ef3cd2b1ed02716bc2e515b8b712b7f6090841e
|
3 |
+
size 347531616
|
preprocessor_config.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_normalize": true,
|
3 |
+
"do_rescale": true,
|
4 |
+
"do_resize": true,
|
5 |
+
"image_mean": [
|
6 |
+
0.485,
|
7 |
+
0.456,
|
8 |
+
0.406
|
9 |
+
],
|
10 |
+
"image_processor_type": "ViTImageProcessor",
|
11 |
+
"image_std": [
|
12 |
+
0.229,
|
13 |
+
0.224,
|
14 |
+
0.225
|
15 |
+
],
|
16 |
+
"resample": 3,
|
17 |
+
"rescale_factor": 0.00392156862745098,
|
18 |
+
"size": {
|
19 |
+
"height": 224,
|
20 |
+
"width": 224
|
21 |
+
}
|
22 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e5ca702b59ed032851ccc2fa57d8ef38a9e73e34a95916e0c8c5d21a62c91976
|
3 |
+
size 4600
|