Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,28 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: cc-by-4.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-4.0
|
| 3 |
+
task_categories:
|
| 4 |
+
- image-classification
|
| 5 |
+
tags:
|
| 6 |
+
- MNIST
|
| 7 |
+
- Image
|
| 8 |
+
pretty_name: Kyrgyz MNIST
|
| 9 |
+
size_categories:
|
| 10 |
+
- 10K<n<100K
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
The data is based on [Kyrgyz MNIST](https://github.com/Akyl-AI/kyrgyz_MNIST).
|
| 14 |
+
It is based on the Test Set:
|
| 15 |
+
|
| 16 |
+
Reproduce by:
|
| 17 |
+
|
| 18 |
+
```python
|
| 19 |
+
numSamplesPerCls = 400
|
| 20 |
+
seedNum = 512
|
| 21 |
+
|
| 22 |
+
dfData = pd.read_csv(r'test.csv')
|
| 23 |
+
dfT = dfData.groupby('label', group_keys = False).sample(n = numSamplesPerCls, replace = False, random_state = seedNum)
|
| 24 |
+
dfT = dfT.reset_index(drop = False)
|
| 25 |
+
dfT = dfT.rename(columns = {'index': 'img_index'})
|
| 26 |
+
|
| 27 |
+
dfT.to_csv(r'MNISTKyrgyzTest400.csv', index = False)
|
| 28 |
+
```
|