Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -105,14 +105,17 @@ Create your virtual environment to help manage dependencies and prevent conflict
|
|
| 105 |
This script is helpful for quick local inspection, prototyping, or lightweight evaluation without downloading the full dataset.
|
| 106 |
|
| 107 |
Usage:
|
|
|
|
| 108 |
python3 utils/sample_swim.py --output-dir ./samples --count 100
|
| 109 |
|
|
|
|
| 110 |
Arguments:
|
| 111 |
--repo-id Hugging Face dataset repository ID
|
| 112 |
--image-subdir Path to image subdirectory inside the dataset repo
|
| 113 |
--label-subdir Path to corresponding label subdirectory
|
| 114 |
--output-dir Directory to save downloaded files
|
| 115 |
--count Number of samples to download
|
|
|
|
| 116 |
|
| 117 |
Example Usage with all args:
|
| 118 |
```
|
|
@@ -135,12 +138,16 @@ Features:
|
|
| 135 |
This script can download the complete dataset for model training or offline access.
|
| 136 |
|
| 137 |
Usage:
|
|
|
|
| 138 |
# Download all chunks (flattened/ YOLO format)
|
|
|
|
| 139 |
python utils/download_swim.py --output-dir ./SWiM --flatten
|
| 140 |
|
|
|
|
| 141 |
# Download specific chunks
|
|
|
|
| 142 |
python3 utils/download_swim.py --chunks 000 001 002 --flatten False
|
| 143 |
-
|
| 144 |
Arguments:
|
| 145 |
--repo-id Hugging Face dataset repository ID
|
| 146 |
--images-parent Parent directory for image chunks (e.g., Baseline/images/train)
|
|
@@ -148,7 +155,7 @@ Arguments:
|
|
| 148 |
--output-dir Where to save the downloaded dataset
|
| 149 |
--flatten Run with "--flatten" to flatten directories. If you want hierarchical chunk folders, omit --flatten
|
| 150 |
--chunks Specific chunk names (e.g., 000 001); omit to download all
|
| 151 |
-
|
| 152 |
Example usage with all args:
|
| 153 |
```
|
| 154 |
python3 utils/download_swim.py
|
|
|
|
| 105 |
This script is helpful for quick local inspection, prototyping, or lightweight evaluation without downloading the full dataset.
|
| 106 |
|
| 107 |
Usage:
|
| 108 |
+
|
| 109 |
python3 utils/sample_swim.py --output-dir ./samples --count 100
|
| 110 |
|
| 111 |
+
```
|
| 112 |
Arguments:
|
| 113 |
--repo-id Hugging Face dataset repository ID
|
| 114 |
--image-subdir Path to image subdirectory inside the dataset repo
|
| 115 |
--label-subdir Path to corresponding label subdirectory
|
| 116 |
--output-dir Directory to save downloaded files
|
| 117 |
--count Number of samples to download
|
| 118 |
+
```
|
| 119 |
|
| 120 |
Example Usage with all args:
|
| 121 |
```
|
|
|
|
| 138 |
This script can download the complete dataset for model training or offline access.
|
| 139 |
|
| 140 |
Usage:
|
| 141 |
+
|
| 142 |
# Download all chunks (flattened/ YOLO format)
|
| 143 |
+
|
| 144 |
python utils/download_swim.py --output-dir ./SWiM --flatten
|
| 145 |
|
| 146 |
+
|
| 147 |
# Download specific chunks
|
| 148 |
+
|
| 149 |
python3 utils/download_swim.py --chunks 000 001 002 --flatten False
|
| 150 |
+
```
|
| 151 |
Arguments:
|
| 152 |
--repo-id Hugging Face dataset repository ID
|
| 153 |
--images-parent Parent directory for image chunks (e.g., Baseline/images/train)
|
|
|
|
| 155 |
--output-dir Where to save the downloaded dataset
|
| 156 |
--flatten Run with "--flatten" to flatten directories. If you want hierarchical chunk folders, omit --flatten
|
| 157 |
--chunks Specific chunk names (e.g., 000 001); omit to download all
|
| 158 |
+
```
|
| 159 |
Example usage with all args:
|
| 160 |
```
|
| 161 |
python3 utils/download_swim.py
|