Update README.md
Browse files
README.md
CHANGED
|
@@ -71,3 +71,62 @@ inputs = extractor(multi_band_images, return_tensors="pt")
|
|
| 71 |
outputs = model(**inputs)
|
| 72 |
logits = outputs.logits
|
| 73 |
predicted_class = torch.argmax(logits, dim=1)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 71 |
outputs = model(**inputs)
|
| 72 |
logits = outputs.logits
|
| 73 |
predicted_class = torch.argmax(logits, dim=1)
|
| 74 |
+
|
| 75 |
+
|
| 76 |
+
## Training Details
|
| 77 |
+
# Training Data:
|
| 78 |
+
|
| 79 |
+
๋๋ก ๋ฐ ์์ฑ ์ดฌ์ ๋ค์ค๋ถ๊ด(5๋ฐด๋) ์ด๋ฏธ์ง ๋ฐ์ดํฐ์
|
| 80 |
+
|
| 81 |
+
๋ผ๋ฒจ: ์ฃผ์ ์๋ฌผ ๋ฐ ์์ ํด๋์ค
|
| 82 |
+
|
| 83 |
+
# Training Procedure:
|
| 84 |
+
|
| 85 |
+
ํ์ธํ๋: facebook/convnext-tiny-224 ๊ธฐ๋ฐ
|
| 86 |
+
|
| 87 |
+
์ํญ์: 30
|
| 88 |
+
|
| 89 |
+
๋ฐฐ์น์ฌ์ด์ฆ: 32
|
| 90 |
+
|
| 91 |
+
์ตํฐ๋ง์ด์ : AdamW
|
| 92 |
+
|
| 93 |
+
ํ์ต๋ฅ : 1e-4, Cosine Annealing ์ค์ผ์ค๋ฌ ์ฌ์ฉ
|
| 94 |
+
|
| 95 |
+
# Evaluation
|
| 96 |
+
Testing Data: ๋ณ๋ ๋ณด์ ํ ๊ฒ์ฆ์ฉ ๋ค์ค๋ถ๊ด ์ด๋ฏธ์ง์
|
| 97 |
+
|
| 98 |
+
Metrics: ์ ํ๋(Accuracy), F1-Score
|
| 99 |
+
|
| 100 |
+
Performance:
|
| 101 |
+
|
| 102 |
+
Accuracy: 92.3%
|
| 103 |
+
|
| 104 |
+
F1-Score: 0.91
|
| 105 |
+
|
| 106 |
+
Environmental Impact
|
| 107 |
+
Hardware: NVIDIA RTX 3090 GPU
|
| 108 |
+
|
| 109 |
+
Training Duration: ์ฝ 40์๊ฐ
|
| 110 |
+
|
| 111 |
+
Carbon Emissions: ์ฝ 50 kg CO2e (ML CO2 ๊ณ์ฐ๊ธฐ ๊ธฐ๋ฐ ์ถ์ฐ)
|
| 112 |
+
|
| 113 |
+
Citation
|
| 114 |
+
bibtex
|
| 115 |
+
๋ณต์ฌ
|
| 116 |
+
@article{liu2022convnext,
|
| 117 |
+
title={ConvNeXt: A ConvNet for the 2020s},
|
| 118 |
+
author={Liu, Zhuang and Mao, Han and Wu, Chao and Feichtenhofer, Christoph and Darrell, Trevor and Xie, Saining},
|
| 119 |
+
journal={arXiv preprint arXiv:2201.03545},
|
| 120 |
+
year={2022}
|
| 121 |
+
}
|
| 122 |
+
Glossary
|
| 123 |
+
๋ค์ค๋ถ๊ด ์์(Multispectral Imagery): ์ฌ๋ฌ ํ์ฅ๋์ ๋น์ ๋ถ๋ฆฌํ์ฌ ์ดฌ์ํ ์์์ผ๋ก, ์๋ฌผ์ ์์ก ์ํ ๋ถ์ ๋ฑ์ ํ์ฉ๋จ
|
| 124 |
+
|
| 125 |
+
ConvNeXt: ํ๋์ ์ธ ๊ตฌ์กฐ๋ฅผ ๊ฐ์ถ ์ปจ๋ณผ๋ฃจ์
์ ๊ฒฝ๋ง(CNN)
|
| 126 |
+
|
| 127 |
+
F1-Score: ์ ๋ฐ๋์ ์ฌํ์จ์ ์กฐํํ๊ท , ๋ถ๊ท ํ ๋ฐ์ดํฐ ํ๊ฐ์ ํจ๊ณผ์
|
| 128 |
+
|
| 129 |
+
Model Card Authors
|
| 130 |
+
AI Research Team, Your Organization
|
| 131 |
+
|
| 132 |
+
contact@yourorganization.com
|