Update readme

README.md

@@ -39,7 +39,7 @@ dataset_info:
   - name: image
     dtype: image
   splits:
-  - name: full
+  - name: test
     num_bytes: 33800694.0
     num_examples: 347
   download_size: 33755238
@@ -67,7 +67,7 @@ dataset_info:
   - name: image
     dtype: image
   splits:
-  - name: full
+  - name: test
     num_bytes: 31270976.0
     num_examples: 345
   download_size: 31240949
@@ -99,7 +99,7 @@ dataset_info:
   - name: image
     dtype: image
   splits:
-  - name: full
+  - name: test
     num_bytes: 22656597.0
     num_examples: 674
   download_size: 22591665
@@ -129,7 +129,7 @@ dataset_info:
   - name: image
     dtype: image
   splits:
-  - name: full
+  - name: test
     num_bytes: 16315782.0
     num_examples: 185
   download_size: 16313529
@@ -153,7 +153,7 @@ dataset_info:
   - name: image
     dtype: image
   splits:
-  - name: full
+  - name: test
     num_bytes: 12072018.0
     num_examples: 194
   download_size: 11900512
@@ -181,7 +181,7 @@ dataset_info:
   - name: image
     dtype: image
   splits:
-  - name: full
+  - name: test
     num_bytes: 12606317.0
     num_examples: 92
   download_size: 12606757
@@ -189,27 +189,27 @@ dataset_info:
 configs:
 - config_name: animals
   data_files:
-  - split: full
+  - split: test
     path: animals/full-*
 - config_name: cars
   data_files:
-  - split: full
+  - split: test
     path: cars/full-*
 - config_name: celebrity
   data_files:
-  - split: full
+  - split: test
     path: celebrity/full-*
 - config_name: plants
   data_files:
-  - split: full
+  - split: test
     path: plants/full-*
 - config_name: products
   data_files:
-  - split: full
+  - split: test
     path: products/full-*
 - config_name: sights
   data_files:
-  - split: full
+  - split: test
     path: sights/full-*
 task_categories:
 - visual-question-answering
@@ -248,7 +248,7 @@ This dataset card aims to be a base template for new datasets. It has been gener
 - **License:** Images are owned by their creators, the collection of data is distributed under cc-by-nc 4.0 license. The data is provided as-is without any warranty or guarantee of fitness for a particular purpose. Please refer to the license file in the repository for more details.
 
 ### Dataset Sources 
-- Images are collected from the Internet with a focus on images NOT featured on Wikipedia, because we expected those to be less likely to be part of existing training datasets.
+- Images are collected from the Internet with a focus on images NOT featured on Wikipedia, because we expected those to be less likely to be part of existing training datasets. Images are filtered and downscaled to be within 400x300 as a minimum (except Celeb1k) and 1280x1024 as a maximum.
 
 - **Paper :** currently under review for a scientific AI conference. Will be published in first half of 2025.