Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -13,7 +13,15 @@ tags:
|
|
| 13 |
- image-classification
|
| 14 |
dataset_summary: '
|
| 15 |
|
| 16 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
|
| 18 |
|
| 19 |
This is a [FiftyOne](https://github.com/voxel51/fiftyone) dataset with 143 samples.
|
|
|
|
| 13 |
- image-classification
|
| 14 |
dataset_summary: '
|
| 15 |
|
| 16 |
+
These photos are used in the FiftyOne getting started webinar. The images have a prediction label where were generated by
|
| 17 |
+
self-supervised classification through a OpenClip Model.
|
| 18 |
+
https://github.com/thesteve0/fiftyone-getting-started/blob/main/5_generating_labels.py
|
| 19 |
+
They were then manually cleaned to produce the ground truth label.
|
| 20 |
+
https://github.com/thesteve0/fiftyone-getting-started/blob/main/6_clean_labels.md
|
| 21 |
+
|
| 22 |
+
They are public domain photos taken from Flickr.
|
| 23 |
+
They will be used to evaluate classification models. They are purely for teaching purposes and are not recommended for any other use.
|
| 24 |
+
No other guarantees provided.
|
| 25 |
|
| 26 |
|
| 27 |
This is a [FiftyOne](https://github.com/voxel51/fiftyone) dataset with 143 samples.
|