Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -32,6 +32,14 @@ A subset from Laion2B (a multimodal dataset), around 143M image-text pairs (only
|
|
| 32 |
- Homepage: [laion-5b](https://laion.ai/blog/laion-5b/)
|
| 33 |
- Huggingface: [laion/laion2B-multi](https://huggingface.co/datasets/laion/laion2B-multi)
|
| 34 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 35 |
## Lisence
|
| 36 |
|
| 37 |
CC-BY-4.0
|
|
|
|
| 32 |
- Homepage: [laion-5b](https://laion.ai/blog/laion-5b/)
|
| 33 |
- Huggingface: [laion/laion2B-multi](https://huggingface.co/datasets/laion/laion2B-multi)
|
| 34 |
|
| 35 |
+
## 下载 Download
|
| 36 |
+
|
| 37 |
+
```bash
|
| 38 |
+
mkdir laion2b_chinese_release && cd laion2b_chinese_release
|
| 39 |
+
for i in {00000..00012}; do wget https://huggingface.co/datasets/IDEA-CCNL/laion2B-multi-chinese-subset/resolve/main/data/train-$i-of-00013.parquet; done
|
| 40 |
+
cd ..
|
| 41 |
+
```
|
| 42 |
+
|
| 43 |
## Lisence
|
| 44 |
|
| 45 |
CC-BY-4.0
|