End of training
Browse files- README.md +134 -0
- config.json +118 -0
- model.safetensors +3 -0
- preprocessor_config.json +22 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,134 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: microsoft/cvt-21-384-22k
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
datasets:
|
7 |
+
- imagefolder
|
8 |
+
metrics:
|
9 |
+
- accuracy
|
10 |
+
model-index:
|
11 |
+
- name: cvt-21-384-22k-finetuned-LeafType
|
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: validation
|
21 |
+
args: default
|
22 |
+
metrics:
|
23 |
+
- name: Accuracy
|
24 |
+
type: accuracy
|
25 |
+
value: 0.9310344827586207
|
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 |
+
# cvt-21-384-22k-finetuned-LeafType
|
32 |
+
|
33 |
+
This model is a fine-tuned version of [microsoft/cvt-21-384-22k](https://huggingface.co/microsoft/cvt-21-384-22k) on the imagefolder dataset.
|
34 |
+
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 0.2232
|
36 |
+
- Accuracy: 0.9310
|
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: 2e-05
|
56 |
+
- train_batch_size: 160
|
57 |
+
- eval_batch_size: 16
|
58 |
+
- seed: 42
|
59 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
60 |
+
- lr_scheduler_type: linear
|
61 |
+
- num_epochs: 60
|
62 |
+
|
63 |
+
### Training results
|
64 |
+
|
65 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
66 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
67 |
+
| No log | 1.0 | 11 | 0.4857 | 0.7586 |
|
68 |
+
| No log | 2.0 | 22 | 0.2275 | 0.9163 |
|
69 |
+
| No log | 3.0 | 33 | 0.2791 | 0.8867 |
|
70 |
+
| No log | 4.0 | 44 | 0.3083 | 0.8522 |
|
71 |
+
| No log | 5.0 | 55 | 0.1496 | 0.9458 |
|
72 |
+
| No log | 6.0 | 66 | 0.2092 | 0.8966 |
|
73 |
+
| No log | 7.0 | 77 | 0.3349 | 0.8522 |
|
74 |
+
| No log | 8.0 | 88 | 0.1513 | 0.9557 |
|
75 |
+
| No log | 9.0 | 99 | 0.1385 | 0.9557 |
|
76 |
+
| No log | 10.0 | 110 | 0.1676 | 0.9360 |
|
77 |
+
| No log | 11.0 | 121 | 0.1284 | 0.9507 |
|
78 |
+
| No log | 12.0 | 132 | 0.1981 | 0.9015 |
|
79 |
+
| No log | 13.0 | 143 | 0.1346 | 0.9360 |
|
80 |
+
| No log | 14.0 | 154 | 0.3235 | 0.8621 |
|
81 |
+
| No log | 15.0 | 165 | 0.1506 | 0.9310 |
|
82 |
+
| No log | 16.0 | 176 | 0.1536 | 0.9360 |
|
83 |
+
| No log | 17.0 | 187 | 0.1529 | 0.9458 |
|
84 |
+
| No log | 18.0 | 198 | 0.2431 | 0.8966 |
|
85 |
+
| No log | 19.0 | 209 | 0.2251 | 0.9163 |
|
86 |
+
| No log | 20.0 | 220 | 0.2068 | 0.9163 |
|
87 |
+
| No log | 21.0 | 231 | 0.1429 | 0.9458 |
|
88 |
+
| No log | 22.0 | 242 | 0.2078 | 0.9163 |
|
89 |
+
| No log | 23.0 | 253 | 0.1428 | 0.9360 |
|
90 |
+
| No log | 24.0 | 264 | 0.2391 | 0.8916 |
|
91 |
+
| No log | 25.0 | 275 | 0.1512 | 0.9458 |
|
92 |
+
| No log | 26.0 | 286 | 0.1512 | 0.9458 |
|
93 |
+
| No log | 27.0 | 297 | 0.1303 | 0.9606 |
|
94 |
+
| No log | 28.0 | 308 | 0.2381 | 0.8867 |
|
95 |
+
| No log | 29.0 | 319 | 0.1544 | 0.9458 |
|
96 |
+
| No log | 30.0 | 330 | 0.2149 | 0.8916 |
|
97 |
+
| No log | 31.0 | 341 | 0.1387 | 0.9557 |
|
98 |
+
| No log | 32.0 | 352 | 0.1698 | 0.9360 |
|
99 |
+
| No log | 33.0 | 363 | 0.1750 | 0.9409 |
|
100 |
+
| No log | 34.0 | 374 | 0.2041 | 0.9113 |
|
101 |
+
| No log | 35.0 | 385 | 0.1888 | 0.9360 |
|
102 |
+
| No log | 36.0 | 396 | 0.1527 | 0.9458 |
|
103 |
+
| No log | 37.0 | 407 | 0.1721 | 0.9360 |
|
104 |
+
| No log | 38.0 | 418 | 0.1720 | 0.9409 |
|
105 |
+
| No log | 39.0 | 429 | 0.1810 | 0.9360 |
|
106 |
+
| No log | 40.0 | 440 | 0.2041 | 0.9261 |
|
107 |
+
| No log | 41.0 | 451 | 0.1945 | 0.9261 |
|
108 |
+
| No log | 42.0 | 462 | 0.2211 | 0.9163 |
|
109 |
+
| No log | 43.0 | 473 | 0.2284 | 0.9163 |
|
110 |
+
| No log | 44.0 | 484 | 0.1905 | 0.9261 |
|
111 |
+
| No log | 45.0 | 495 | 0.2329 | 0.9113 |
|
112 |
+
| 0.2716 | 46.0 | 506 | 0.1839 | 0.9360 |
|
113 |
+
| 0.2716 | 47.0 | 517 | 0.1854 | 0.9310 |
|
114 |
+
| 0.2716 | 48.0 | 528 | 0.1923 | 0.9310 |
|
115 |
+
| 0.2716 | 49.0 | 539 | 0.2254 | 0.9261 |
|
116 |
+
| 0.2716 | 50.0 | 550 | 0.2677 | 0.9015 |
|
117 |
+
| 0.2716 | 51.0 | 561 | 0.2620 | 0.9015 |
|
118 |
+
| 0.2716 | 52.0 | 572 | 0.2224 | 0.9261 |
|
119 |
+
| 0.2716 | 53.0 | 583 | 0.2238 | 0.9310 |
|
120 |
+
| 0.2716 | 54.0 | 594 | 0.2355 | 0.9212 |
|
121 |
+
| 0.2716 | 55.0 | 605 | 0.2533 | 0.9261 |
|
122 |
+
| 0.2716 | 56.0 | 616 | 0.2237 | 0.9310 |
|
123 |
+
| 0.2716 | 57.0 | 627 | 0.2296 | 0.9310 |
|
124 |
+
| 0.2716 | 58.0 | 638 | 0.2361 | 0.9310 |
|
125 |
+
| 0.2716 | 59.0 | 649 | 0.2358 | 0.9310 |
|
126 |
+
| 0.2716 | 60.0 | 660 | 0.2232 | 0.9310 |
|
127 |
+
|
128 |
+
|
129 |
+
### Framework versions
|
130 |
+
|
131 |
+
- Transformers 4.38.1
|
132 |
+
- Pytorch 1.10.0+cu111
|
133 |
+
- Datasets 2.17.1
|
134 |
+
- Tokenizers 0.15.2
|
config.json
ADDED
@@ -0,0 +1,118 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "microsoft/cvt-21-384-22k",
|
3 |
+
"architectures": [
|
4 |
+
"CvtForImageClassification"
|
5 |
+
],
|
6 |
+
"attention_drop_rate": [
|
7 |
+
0.0,
|
8 |
+
0.0,
|
9 |
+
0.0
|
10 |
+
],
|
11 |
+
"cls_token": [
|
12 |
+
false,
|
13 |
+
false,
|
14 |
+
true
|
15 |
+
],
|
16 |
+
"depth": [
|
17 |
+
1,
|
18 |
+
4,
|
19 |
+
16
|
20 |
+
],
|
21 |
+
"drop_path_rate": [
|
22 |
+
0.0,
|
23 |
+
0.0,
|
24 |
+
0.1
|
25 |
+
],
|
26 |
+
"drop_rate": [
|
27 |
+
0.0,
|
28 |
+
0.0,
|
29 |
+
0.0
|
30 |
+
],
|
31 |
+
"embed_dim": [
|
32 |
+
64,
|
33 |
+
192,
|
34 |
+
384
|
35 |
+
],
|
36 |
+
"id2label": {
|
37 |
+
"0": "leaflet",
|
38 |
+
"1": "pinnate"
|
39 |
+
},
|
40 |
+
"image_size": 384,
|
41 |
+
"initializer_range": 0.02,
|
42 |
+
"kernel_qkv": [
|
43 |
+
3,
|
44 |
+
3,
|
45 |
+
3
|
46 |
+
],
|
47 |
+
"label2id": {
|
48 |
+
"leaflet": 0,
|
49 |
+
"pinnate": 1
|
50 |
+
},
|
51 |
+
"layer_norm_eps": 1e-12,
|
52 |
+
"mlp_ratio": [
|
53 |
+
4.0,
|
54 |
+
4.0,
|
55 |
+
4.0
|
56 |
+
],
|
57 |
+
"model_type": "cvt",
|
58 |
+
"num_channels": 3,
|
59 |
+
"num_heads": [
|
60 |
+
1,
|
61 |
+
3,
|
62 |
+
6
|
63 |
+
],
|
64 |
+
"num_stages": 3,
|
65 |
+
"padding_kv": [
|
66 |
+
1,
|
67 |
+
1,
|
68 |
+
1
|
69 |
+
],
|
70 |
+
"padding_q": [
|
71 |
+
1,
|
72 |
+
1,
|
73 |
+
1
|
74 |
+
],
|
75 |
+
"patch_padding": [
|
76 |
+
2,
|
77 |
+
1,
|
78 |
+
1
|
79 |
+
],
|
80 |
+
"patch_sizes": [
|
81 |
+
7,
|
82 |
+
3,
|
83 |
+
3
|
84 |
+
],
|
85 |
+
"patch_stride": [
|
86 |
+
4,
|
87 |
+
2,
|
88 |
+
2
|
89 |
+
],
|
90 |
+
"pos_embed": [
|
91 |
+
false,
|
92 |
+
false,
|
93 |
+
false
|
94 |
+
],
|
95 |
+
"problem_type": "single_label_classification",
|
96 |
+
"qkv_bias": [
|
97 |
+
true,
|
98 |
+
true,
|
99 |
+
true
|
100 |
+
],
|
101 |
+
"qkv_projection_method": [
|
102 |
+
"dw_bn",
|
103 |
+
"dw_bn",
|
104 |
+
"dw_bn"
|
105 |
+
],
|
106 |
+
"stride_kv": [
|
107 |
+
2,
|
108 |
+
2,
|
109 |
+
2
|
110 |
+
],
|
111 |
+
"stride_q": [
|
112 |
+
1,
|
113 |
+
1,
|
114 |
+
1
|
115 |
+
],
|
116 |
+
"torch_dtype": "float32",
|
117 |
+
"transformers_version": "4.38.1"
|
118 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:789b6fd135b9f26be242846507b8dd167f608c637281463a6629a68416eb26c1
|
3 |
+
size 125239600
|
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 |
+
0.5,
|
7 |
+
0.5,
|
8 |
+
0.5
|
9 |
+
],
|
10 |
+
"image_processor_type": "ViTImageProcessor",
|
11 |
+
"image_std": [
|
12 |
+
0.5,
|
13 |
+
0.5,
|
14 |
+
0.5
|
15 |
+
],
|
16 |
+
"resample": 2,
|
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 sha256:18f0a47968d1374034635bd833a95d2720d15d36abb0e6587c8e40ed24806264
|
3 |
+
size 4399
|