Update upload_captions.py
Browse files- upload_captions.py +16 -4
upload_captions.py
CHANGED
@@ -1,3 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
import os
|
2 |
import json
|
3 |
from datasets import Dataset, Features, Value, Image
|
@@ -102,18 +113,19 @@ def create_and_push_dataset(data, dataset_name, token):
|
|
102 |
# 推送到Hub
|
103 |
dataset.push_to_hub(
|
104 |
dataset_name,
|
105 |
-
private=
|
106 |
)
|
107 |
|
108 |
print(f"数据集成功上传到: https://huggingface.co/datasets/{dataset_name}")
|
109 |
|
110 |
if __name__ == "__main__":
|
111 |
# 配置
|
112 |
-
IMAGE_CAPTION_DIR = "/
|
113 |
-
DATASET_NAME = "housearch/
|
114 |
|
115 |
# 从https://huggingface.co/settings/tokens获取token
|
116 |
-
HF_TOKEN = "填入你的token"
|
|
|
117 |
|
118 |
# 加载图像和标题
|
119 |
print(f"从 {IMAGE_CAPTION_DIR} 加载图像和标题...")
|
|
|
1 |
+
"""
|
2 |
+
1.創建一個data的資料夾,把圖檔和caption的json檔案上傳到data
|
3 |
+
2.上傳 upload_captions.py 腳本檔
|
4 |
+
3.填入你的金鑰
|
5 |
+
4.在terminal中執行以下兩行指令
|
6 |
+
|
7 |
+
pip install datasets
|
8 |
+
python upload_captions.py
|
9 |
+
|
10 |
+
"""
|
11 |
+
|
12 |
import os
|
13 |
import json
|
14 |
from datasets import Dataset, Features, Value, Image
|
|
|
113 |
# 推送到Hub
|
114 |
dataset.push_to_hub(
|
115 |
dataset_name,
|
116 |
+
private=False # 如果您希望数据集是公开的,设置为False
|
117 |
)
|
118 |
|
119 |
print(f"数据集成功上传到: https://huggingface.co/datasets/{dataset_name}")
|
120 |
|
121 |
if __name__ == "__main__":
|
122 |
# 配置
|
123 |
+
IMAGE_CAPTION_DIR = "/workspace/data"
|
124 |
+
DATASET_NAME = "housearch/Park-PFI" # 从您的错误消息中看到的数据集名称
|
125 |
|
126 |
# 从https://huggingface.co/settings/tokens获取token
|
127 |
+
#HF_TOKEN = "填入你的token"
|
128 |
+
HF_TOKEN = "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"
|
129 |
|
130 |
# 加载图像和标题
|
131 |
print(f"从 {IMAGE_CAPTION_DIR} 加载图像和标题...")
|