Update chess_ground-targz.py

chess_ground-targz.py

@@ -3,7 +3,7 @@ import datasets
 
 # Description of the dataset
 _DESCRIPTION = """\
-Dataset for extracting notations from chess-scoresheets.
+Dataset for extracting notations from chess scoresheets.
 """
 
 # BibTeX citation for the dataset
@@ -18,45 +18,72 @@ year={2020}
 # License of the dataset
 _LICENSE = "Creative Commons Attribution 3.0"
 
-# Class for loading the chess ground truth dataset
-class ChessGroundTargz(datasets.GeneratorBasedBuilder):
-    """Chess Moves ground truths for chess scoresheets"""
-    
+
+class ChessImageTextDataset(datasets.GeneratorBasedBuilder):
+    """Dataset for linking chess scoresheets images and their ground truth text."""
+
     def _info(self):
-        # Define the features of your dataset (i.e., the columns)
+        # Define the features of your dataset (image and text)
         features = datasets.Features(
             {
-                "text": datasets.Value("string"),
+                "image": datasets.Image(),  # Image feature for chess scoresheets
+                "text": datasets.Value("string"),  # Text feature for chess notations
             }
         )
         return datasets.DatasetInfo(
             description=_DESCRIPTION,
             features=features,
-            homepage="Chesscorner/jsonl-chess-dataset",  # Update with the correct homepage link
+            homepage="Chesscorner/chess-images",  # Update with the correct homepage link
             license=_LICENSE,
             citation=_CITATION,
         )
 
     def _split_generators(self, dl_manager):
         """Define the splits of the dataset."""
-        # Instead of downloading, we are now referencing the dataset on Hugging Face
-        # Hugging Face handles downloading and caching automatically
+        
+        # Load the image dataset
+        image_dataset_url = "https://huggingface.co/datasets/Chesscorner/chess-images/resolve/main/images.tar.gz"
+        extracted_image_path = dl_manager.download_and_extract(image_dataset_url)
+
+        # Load the text dataset (ground truths)
+        text_dataset_url = "https://huggingface.co/datasets/Chesscorner/jsonl-chess-dataset/resolve/main/train.jsonl/train.jsonl"
+        text_filepath = dl_manager.download_and_extract(text_dataset_url)
+
         return [
             datasets.SplitGenerator(
                 name=datasets.Split.TRAIN,
                 gen_kwargs={
-                    "filepath": dl_manager.download_and_extract("https://huggingface.co/datasets/Chesscorner/jsonl-chess-dataset/resolve/main/train.jsonl/train.jsonl"),
+                    "image_path": extracted_image_path,
+                    "text_filepath": text_filepath,
                 },
             ),
         ]
 
-    def _generate_examples(self, filepath):
+    def _generate_examples(self, image_path, text_filepath):
+        """Generate examples linking images to text."""
+
         idx = 0
-        # open the file and read the lines
-        with open(filepath, encoding="utf-8") as fp:
+        # Load the text dataset
+        with open(text_filepath, encoding="utf-8") as fp:
             for line in fp:
-                # load json line
                 obj = json.loads(line)
-                yield idx, obj
-                idx += 1 
+                text = obj["text"]
+                
+                # Extract image ID from the text (assuming the first 3 characters of text match image filename)
+                image_id = text[:3]  # Adjust this logic based on the exact pattern you have
+
+                # Find the corresponding image
+                image_file = f"{image_path}/{image_id}.png"  # Adjust file extension if needed
+                
+                # Check if the image file exists (you can add extra checks here)
+                if os.path.exists(image_file):
+                    yield idx, {
+                        "image": image_file,
+                        "text": text,
+                    }
+                else:
+                    print(f"Image not found for ID: {image_id}")
+                
+                idx += 1
+