split img paths

NLVR2.py

@@ -44,7 +44,7 @@ _SPLITS = {
     # "test1": "test1.json",
     # "test2": "test2.json",
 }
-_JZ_IMG_FOLDER_PATH = f"{os.environ['cnw_ALL_CCFRSCRATCH']}/local_datasets/nlvr2"
+_JZ_IMG_FOLDER_PATH = f"{os.environ['cnw_ALL_CCFRSCRATCH']}/local_datasets/nlvr2/"
 _IMG_SPLITS = {
     # "train": "train_nlvr2.zip",
     "validation": "dev_nlvr2.zip",
@@ -89,26 +89,26 @@ class NLVR2Dataset(datasets.GeneratorBasedBuilder):
         files_path = dl_manager.download_and_extract(urls)
 
         images_files = {
-            "Default": {
-                # "train": os.path.join(_JZ_IMG_FOLDER_PATH, _IMG_SPLITS["train"]),
-                "validation": os.path.join(_JZ_IMG_FOLDER_PATH, _IMG_SPLITS["validation"]),
-                # "test1": os.path.join(_JZ_IMG_FOLDER_PATH, _IMG_SPLITS["test1"]),
-                # "test2": os.path.join(_JZ_IMG_FOLDER_PATH, _IMG_SPLITS["test2"]),
-            },
+            # "train": os.path.join(_JZ_IMG_FOLDER_PATH, _IMG_SPLITS["train"]),
+            "validation": os.path.join(_JZ_IMG_FOLDER_PATH, _IMG_SPLITS["validation"]),
+            # "test1": os.path.join(_JZ_IMG_FOLDER_PATH, _IMG_SPLITS["test1"]),
+            # "test2": os.path.join(_JZ_IMG_FOLDER_PATH, _IMG_SPLITS["test2"]),
         }
-        images_path = dl_manager.download_and_extract(images_files)
+        train_img_path = os.path.join(dl_manager.extract(images_files["train"]),  "train_nlvr2")
+        validation_img_path = os.path.join(dl_manager.extract(images_files["validation"]),  "dev_nlvr2")
+        test1_img_path = os.path.join(dl_manager.extract(images_files["test1"]),  "test1_nlvr2")
         return [
             # datasets.SplitGenerator(
             #     name=datasets.Split.TRAIN,
-            #     gen_kwargs={"files_paths": [files_path[self.config.name]["train"]], "images_path": images_path, "split_name": "train"},
+            #     gen_kwargs={"files_paths": [files_path[self.config.name]["train"]], "images_path": train_img_path, "split_name": "train"},
             # ),
             datasets.SplitGenerator(
                 name=datasets.Split.VALIDATION,
-                gen_kwargs={"files_paths": [files_path[self.config.name]["validation"]], "images_path": images_path, "split_name": "dev"},
+                gen_kwargs={"files_paths": [files_path[self.config.name]["validation"]], "images_path": validation_img_path, "split_name": "dev"},
             ),
             # datasets.SplitGenerator(
             #     name=datasets.Split.TEST,
-            #     gen_kwargs={"files_paths": [files_path[self.config.name]["test1"], files_path[self.config.name]["test2"]], "images_path": images_path, "split_name": "test1"},
+            #     gen_kwargs={"files_paths": [files_path[self.config.name]["test1"], files_path[self.config.name]["test2"]], "images_path": test1_img_path, "split_name": "test1"},
             # ),
         ]