Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -85,6 +85,21 @@ Then make sure that Docker Compose is running, navigate to `examples/pipelines/f
|
|
85 |
fondant run pipeline --local
|
86 |
```
|
87 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
88 |
To visually inspect the results quickly, you can use:
|
89 |
|
90 |
```bash
|
|
|
85 |
fondant run pipeline --local
|
86 |
```
|
87 |
|
88 |
+
**Note:** For local testing purposes, the pipeline will only download the first 100,000 images.
|
89 |
+
If you want to download the full dataset, you will need to modify the component arguments in the `pipeline.py` file,
|
90 |
+
specifically the following part:
|
91 |
+
|
92 |
+
```
|
93 |
+
load_from_hf_hub = ComponentOp(
|
94 |
+
component_dir="components/load_from_hf_hub",
|
95 |
+
arguments={
|
96 |
+
"dataset_name": "fondant-ai/fondant-cc-25m",
|
97 |
+
"column_name_mapping": load_component_column_mapping,
|
98 |
+
"n_rows_to_load": <HERE INSERT THE NUMBER OF IMAGES YOU WANT TO DOWNLOAD>
|
99 |
+
},
|
100 |
+
)
|
101 |
+
```
|
102 |
+
|
103 |
To visually inspect the results quickly, you can use:
|
104 |
|
105 |
```bash
|