khanitachi commited on
Commit
d813871
·
verified ·
1 Parent(s): b5ea811

Fix format

Browse files
Files changed (1) hide show
  1. README.md +6 -4
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
- ## Slices
59
 
60
- The dataset is split into a single training split (`train`).
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.