Update SLR.py

SLR.py

@@ -68,11 +68,11 @@ class SLR(datasets.GeneratorBasedBuilder):
                 {
                     "id": datasets.Value("string"),
                     "words": datasets.Sequence(datasets.Value("string")),
-                    "tokens": datasets.Sequence(datasets.Value("string")),
+                    #"tokens": datasets.Sequence(datasets.Value("string")),
                     "bboxes": datasets.Sequence(datasets.Sequence(datasets.Value("int64"))),
                     "ner_tags": datasets.Sequence(
                         datasets.features.ClassLabel(
-                            names=["Borrower", "BorrowerValue", "City", "CityValue", "PUD","PerMonth","PerYear","State","StateValue"]
+                            names=["Property","PropertyValue","Borrower","BorrowerValue","PUD","PUDMonth","Table","TableValue"]
                         )
                     ),
                     "image": datasets.Array3D(shape=(3, 224, 224), dtype="uint8"),
@@ -102,7 +102,7 @@ class SLR(datasets.GeneratorBasedBuilder):
         img_dir = os.path.join(filepath, "images")
         for guid, file in enumerate(sorted(os.listdir(ann_dir))):
             words=[]
-            tokens = []
+           
             bboxes = []
             ner_tags = []
 
@@ -112,13 +112,14 @@ class SLR(datasets.GeneratorBasedBuilder):
             image_path = os.path.join(img_dir, file)
             image_path = image_path.replace("json", "jpg")
             image, size = load_image(image_path)
-            for item in data['categories']:
+            for item in data['form']:
                 
-                labels=item['name']
-                words.append(labels)
+                labels=item['label']
+                word=item['text']
                 ner_tags.append(labels)
-            for item in data['annotations']:
+                words.append(word)
+            #for item in data['annotations']:
                #bbox=item['bbox']
-               bboxes.append(normalize_bbox(item['bbox'],size))    
+               bboxes.append(normalize_bbox(item['box'],size))    
 
-            yield guid, {"id": str(guid), "words": words , "tokens": tokens, "bboxes": bboxes, "ner_tags": ner_tags, "image_path": image_path, "image": image}
+            yield guid, {"id": str(guid), "words": words , "bboxes": bboxes, "ner_tags": ner_tags, "image_path": image_path, "image": image}