Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -9,11 +9,24 @@ dataset_info:
|
|
9 |
dtype: image
|
10 |
splits:
|
11 |
- name: train
|
12 |
-
num_bytes: 3257571330
|
13 |
num_examples: 5000
|
14 |
download_size: 3296272284
|
15 |
-
dataset_size: 3257571330
|
|
|
|
|
16 |
---
|
17 |
-
#
|
18 |
|
19 |
-
[
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
dtype: image
|
10 |
splits:
|
11 |
- name: train
|
12 |
+
num_bytes: 3257571330
|
13 |
num_examples: 5000
|
14 |
download_size: 3296272284
|
15 |
+
dataset_size: 3257571330
|
16 |
+
size_categories:
|
17 |
+
- 1K<n<10K
|
18 |
---
|
19 |
+
# Instruction-prompted cartoonization dataset
|
20 |
|
21 |
+
This dataset was created from 5000 images randomly sampled from the [Imagenette dataset](https://github.com/fastai/imagenette). For more
|
22 |
+
details on how the dataset was created, check out [this directory](https://github.com/sayakpaul/instruction-tuned-sd/tree/main/data_preparation).
|
23 |
+
|
24 |
+
## Known limitations and biases
|
25 |
+
|
26 |
+
The dataset was derived from Imagenette, which, in turn, was derived from [ImageNet](https://www.image-net.org/). So, naturally, this
|
27 |
+
dataset inherits the limitations and biases of ImageNet.
|
28 |
+
|
29 |
+
## Licensing
|
30 |
+
|
31 |
+
The dataset was derived from Imagenette, which, in turn, was derived from [ImageNet](https://www.image-net.org/). So, this dataset's license
|
32 |
+
is same as ImageNet.
|