Datasets:
Fix format
Browse files
README.md
CHANGED
@@ -39,7 +39,7 @@ This dataset is intended for educational and research purposes, including:
|
|
39 |
* **Image classification:** Train models to recognize different One Piece characters.
|
40 |
* **Image generation:** Fine-tune text-to-image models to generate images of One Piece characters.
|
41 |
|
42 |
-
## Format
|
43 |
|
44 |
The dataset is organized as follows:
|
45 |
|
@@ -48,17 +48,19 @@ The dataset is organized as follows:
|
|
48 |
* **Metadata File:** `metadata.jsonl`
|
49 |
* Contains metadata for each image, with each line being a JSON object containing:
|
50 |
* `file_name`: The name of the image file. (type: string)
|
51 |
-
* `text`: A caption or description of the image. (type: string)
|
52 |
|
53 |
## Features
|
54 |
|
55 |
* **image (jpg):** The image of the One Piece character. (expected type: PIL Image, JPG format)
|
56 |
* **text (str):** A caption or description of the image. (expected type: string)
|
57 |
|
58 |
-
|
59 |
|
60 |
-
|
61 |
|
|
|
|
|
62 |
## Acknowledgements
|
63 |
|
64 |
This dataset uses images primarily from Google Images. Used `simple_image_download` for downloading images using character queries.
|
|
|
39 |
* **Image classification:** Train models to recognize different One Piece characters.
|
40 |
* **Image generation:** Fine-tune text-to-image models to generate images of One Piece characters.
|
41 |
|
42 |
+
<!-- ## Format
|
43 |
|
44 |
The dataset is organized as follows:
|
45 |
|
|
|
48 |
* **Metadata File:** `metadata.jsonl`
|
49 |
* Contains metadata for each image, with each line being a JSON object containing:
|
50 |
* `file_name`: The name of the image file. (type: string)
|
51 |
+
* `text`: A caption or description of the image. (type: string) -->
|
52 |
|
53 |
## Features
|
54 |
|
55 |
* **image (jpg):** The image of the One Piece character. (expected type: PIL Image, JPG format)
|
56 |
* **text (str):** A caption or description of the image. (expected type: string)
|
57 |
|
58 |
+
#TODO: Add image samples
|
59 |
|
60 |
+
<!-- ## Slices
|
61 |
|
62 |
+
The dataset is split into a single training split (`train`).
|
63 |
+
-->
|
64 |
## Acknowledgements
|
65 |
|
66 |
This dataset uses images primarily from Google Images. Used `simple_image_download` for downloading images using character queries.
|