Hugging Face's logo

SodaXII
/
convnextv2-base-1k-224_rice-leaf-disease-augmented-v4_v5_pft

Image Classification
Transformers
TensorBoard
Safetensors
convnextv2
Generated from Trainer
Model card Files
xet
 Metrics Community
SodaXII commited on
Commit
6242e4b
·
verified ·
1 Parent(s): bdd4c37

Model save

Browse files
Files changed (7) hide show
  1. README.md +122 -0
  2. config.json +63 -0
  3. logs/events.out.tfevents.1745554756.c6bc1958280e.1138.0 +3 -0
  4. model.safetensors +3 -0
  5. preprocessor_config.json +22 -0
  6. training_args.bin +3 -0
  7. training_metrics.xlsx +0 -0
README.md ADDED
@@ -0,0 +1,122 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: apache-2.0
4
+ base_model: facebook/convnextv2-base-1k-224
5
+ tags:
6
+ - generated_from_trainer
7
+ metrics:
8
+ - accuracy
9
+ model-index:
10
+ - name: convnextv2-base-1k-224_rice-leaf-disease-augmented-v4_v5_pft
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # convnextv2-base-1k-224_rice-leaf-disease-augmented-v4_v5_pft
18
+
19
+ This model is a fine-tuned version of [facebook/convnextv2-base-1k-224](https://huggingface.co/facebook/convnextv2-base-1k-224) on an unknown dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 0.6774
22
+ - Accuracy: 0.7819
23
+
24
+ ## Model description
25
+
26
+ More information needed
27
+
28
+ ## Intended uses & limitations
29
+
30
+ More information needed
31
+
32
+ ## Training and evaluation data
33
+
34
+ More information needed
35
+
36
+ ## Training procedure
37
+
38
+ ### Training hyperparameters
39
+
40
+ The following hyperparameters were used during training:
41
+ - learning_rate: 0.0003
42
+ - train_batch_size: 64
43
+ - eval_batch_size: 64
44
+ - seed: 42
45
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
46
+ - lr_scheduler_type: cosine_with_restarts
47
+ - lr_scheduler_warmup_steps: 256
48
+ - num_epochs: 30
49
+ - mixed_precision_training: Native AMP
50
+
51
+ ### Training results
52
+
53
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
54
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
55
+ | 2.0691 | 0.5 | 64 | 2.0083 | 0.3523 |
56
+ | 1.9566 | 1.0 | 128 | 1.8732 | 0.5201 |
57
+ | 1.7715 | 1.5 | 192 | 1.6903 | 0.5638 |
58
+ | 1.5752 | 2.0 | 256 | 1.5040 | 0.6074 |
59
+ | 1.4088 | 2.5 | 320 | 1.3569 | 0.6208 |
60
+ | 1.2927 | 3.0 | 384 | 1.2600 | 0.6309 |
61
+ | 1.2173 | 3.5 | 448 | 1.1948 | 0.6577 |
62
+ | 1.1515 | 4.0 | 512 | 1.1464 | 0.6644 |
63
+ | 1.1183 | 4.5 | 576 | 1.1160 | 0.6711 |
64
+ | 1.0893 | 5.0 | 640 | 1.1001 | 0.6879 |
65
+ | 1.0792 | 5.5 | 704 | 1.0898 | 0.6913 |
66
+ | 1.0627 | 6.0 | 768 | 1.0831 | 0.6846 |
67
+ | 1.0714 | 6.5 | 832 | 1.0817 | 0.6846 |
68
+ | 1.0459 | 7.0 | 896 | 1.0483 | 0.6913 |
69
+ | 1.0282 | 7.5 | 960 | 1.0047 | 0.6980 |
70
+ | 0.9605 | 8.0 | 1024 | 0.9774 | 0.7081 |
71
+ | 0.9405 | 8.5 | 1088 | 0.9489 | 0.7114 |
72
+ | 0.9316 | 9.0 | 1152 | 0.9353 | 0.7148 |
73
+ | 0.9174 | 9.5 | 1216 | 0.9208 | 0.7181 |
74
+ | 0.8924 | 10.0 | 1280 | 0.9137 | 0.7215 |
75
+ | 0.9009 | 10.5 | 1344 | 0.9101 | 0.7282 |
76
+ | 0.8844 | 11.0 | 1408 | 0.9092 | 0.7248 |
77
+ | 0.8873 | 11.5 | 1472 | 0.9076 | 0.7215 |
78
+ | 0.8751 | 12.0 | 1536 | 0.8721 | 0.7383 |
79
+ | 0.8553 | 12.5 | 1600 | 0.8617 | 0.7248 |
80
+ | 0.8265 | 13.0 | 1664 | 0.8428 | 0.7416 |
81
+ | 0.8133 | 13.5 | 1728 | 0.8302 | 0.7416 |
82
+ | 0.808 | 14.0 | 1792 | 0.8232 | 0.7483 |
83
+ | 0.7915 | 14.5 | 1856 | 0.8187 | 0.7450 |
84
+ | 0.7975 | 15.0 | 1920 | 0.8157 | 0.7450 |
85
+ | 0.7765 | 15.5 | 1984 | 0.8143 | 0.7450 |
86
+ | 0.8017 | 16.0 | 2048 | 0.8142 | 0.7450 |
87
+ | 0.793 | 16.5 | 2112 | 0.7970 | 0.7584 |
88
+ | 0.7567 | 17.0 | 2176 | 0.7901 | 0.7550 |
89
+ | 0.7576 | 17.5 | 2240 | 0.7785 | 0.7483 |
90
+ | 0.7377 | 18.0 | 2304 | 0.7651 | 0.7651 |
91
+ | 0.7311 | 18.5 | 2368 | 0.7588 | 0.7651 |
92
+ | 0.7276 | 19.0 | 2432 | 0.7566 | 0.7651 |
93
+ | 0.7237 | 19.5 | 2496 | 0.7567 | 0.7651 |
94
+ | 0.7171 | 20.0 | 2560 | 0.7534 | 0.7685 |
95
+ | 0.7158 | 20.5 | 2624 | 0.7529 | 0.7685 |
96
+ | 0.7188 | 21.0 | 2688 | 0.7486 | 0.7651 |
97
+ | 0.7112 | 21.5 | 2752 | 0.7340 | 0.7752 |
98
+ | 0.6912 | 22.0 | 2816 | 0.7297 | 0.7752 |
99
+ | 0.6784 | 22.5 | 2880 | 0.7229 | 0.7785 |
100
+ | 0.6868 | 23.0 | 2944 | 0.7152 | 0.7752 |
101
+ | 0.6701 | 23.5 | 3008 | 0.7132 | 0.7819 |
102
+ | 0.6718 | 24.0 | 3072 | 0.7111 | 0.7752 |
103
+ | 0.671 | 24.5 | 3136 | 0.7105 | 0.7785 |
104
+ | 0.6609 | 25.0 | 3200 | 0.7097 | 0.7785 |
105
+ | 0.6722 | 25.5 | 3264 | 0.7066 | 0.7785 |
106
+ | 0.6526 | 26.0 | 3328 | 0.6959 | 0.7785 |
107
+ | 0.6448 | 26.5 | 3392 | 0.6920 | 0.7919 |
108
+ | 0.6493 | 27.0 | 3456 | 0.6903 | 0.7785 |
109
+ | 0.6394 | 27.5 | 3520 | 0.6816 | 0.7785 |
110
+ | 0.6274 | 28.0 | 3584 | 0.6819 | 0.7819 |
111
+ | 0.6198 | 28.5 | 3648 | 0.6784 | 0.7819 |
112
+ | 0.632 | 29.0 | 3712 | 0.6778 | 0.7819 |
113
+ | 0.634 | 29.5 | 3776 | 0.6776 | 0.7819 |
114
+ | 0.612 | 30.0 | 3840 | 0.6774 | 0.7819 |
115
+
116
+
117
+ ### Framework versions
118
+
119
+ - Transformers 4.48.3
120
+ - Pytorch 2.5.1+cu124
121
+ - Datasets 3.3.2
122
+ - Tokenizers 0.21.1
config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/convnextv2-base-1k-224",
3
+ "architectures": [
4
+ "ConvNextV2ForImageClassification"
5
+ ],
6
+ "depths": [
7
+ 3,
8
+ 3,
9
+ 27,
10
+ 3
11
+ ],
12
+ "drop_path_rate": 0.0,
13
+ "hidden_act": "gelu",
14
+ "hidden_sizes": [
15
+ 128,
16
+ 256,
17
+ 512,
18
+ 1024
19
+ ],
20
+ "id2label": {
21
+ "0": "Bacterial Leaf Blight",
22
+ "1": "Brown Spot",
23
+ "2": "Healthy Rice Leaf",
24
+ "3": "Leaf Blast",
25
+ "4": "Leaf scald",
26
+ "5": "Narrow Brown Leaf Spot",
27
+ "6": "Rice Hispa",
28
+ "7": "Sheath Blight"
29
+ },
30
+ "image_size": 224,
31
+ "initializer_range": 0.02,
32
+ "label2id": {
33
+ "Bacterial Leaf Blight": 0,
34
+ "Brown Spot": 1,
35
+ "Healthy Rice Leaf": 2,
36
+ "Leaf Blast": 3,
37
+ "Leaf scald": 4,
38
+ "Narrow Brown Leaf Spot": 5,
39
+ "Rice Hispa": 6,
40
+ "Sheath Blight": 7
41
+ },
42
+ "layer_norm_eps": 1e-12,
43
+ "model_type": "convnextv2",
44
+ "num_channels": 3,
45
+ "num_stages": 4,
46
+ "out_features": [
47
+ "stage4"
48
+ ],
49
+ "out_indices": [
50
+ 4
51
+ ],
52
+ "patch_size": 4,
53
+ "problem_type": "single_label_classification",
54
+ "stage_names": [
55
+ "stem",
56
+ "stage1",
57
+ "stage2",
58
+ "stage3",
59
+ "stage4"
60
+ ],
61
+ "torch_dtype": "float32",
62
+ "transformers_version": "4.48.3"
63
+ }
logs/events.out.tfevents.1745554756.c6bc1958280e.1138.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:463c0c5fbcbe4a4acdb4e61e6432de19a78294f651531b8c65e4fa46ba457543
3
+ size 44007
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebc571ca38433d131a5ca7646997b1e1e6fbd5e49b46f2ca0c63d1b417afad82
3
+ size 350850048
preprocessor_config.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_pct": 0.875,
3
+ "do_normalize": true,
4
+ "do_rescale": true,
5
+ "do_resize": true,
6
+ "image_mean": [
7
+ 0.485,
8
+ 0.456,
9
+ 0.406
10
+ ],
11
+ "image_processor_type": "ConvNextImageProcessor",
12
+ "image_std": [
13
+ 0.229,
14
+ 0.224,
15
+ 0.225
16
+ ],
17
+ "resample": 3,
18
+ "rescale_factor": 0.00392156862745098,
19
+ "size": {
20
+ "shortest_edge": 224
21
+ }
22
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37f36e97d21e9708911a588db16f86194317d2c15c60c3f56b2ac19458e9a12d
3
+ size 5496
training_metrics.xlsx ADDED
Binary file (11.1 kB). View file