Add filtered rico version where bad hierarchies are removed

captions_hierarchies_images_filtered.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df7c2d2cdb2f5e4a7f3409af1c1e40a7a66549a691fa61b469bf838c2084d53d
+size 2135916754

filter.py

@@ -0,0 +1,21 @@
+import shutil
+
+from pathlib import Path
+from tqdm.auto import tqdm
+
+# filtering file from https://github.com/google-research/google-research/tree/master/seq2act/data_generation
+with open("./filter_24598.txt", "r") as f:
+    filtered_files = f.read().replace(".json", "").split("\n")[:-1]
+
+print("num_filtered:", len(filtered_files))
+
+# make filtered directories
+Path("captions_filtered").mkdir(exist_ok=True)
+Path("hierarchies_filtered").mkdir(exist_ok=True)
+Path("images_filtered").mkdir(exist_ok=True)
+
+# copy filtered files into filtered directories
+for file in tqdm(filtered_files):
+    shutil.copy(f"captions/{file}.txt", "captions_filtered")
+    shutil.copy(f"hierarchies/{file}.json", "hierarchies_filtered")
+    shutil.copy(f"images/{file}.jpg", "images_filtered")

rico_captions.py

@@ -51,6 +51,7 @@ _LICENSE = ""
 # This can be an arbitrary nested dict/list of URLs (see below in `_split_generators` method)
 _DATA_URLs = {
     "screenshots_captions": "https://huggingface.co/datasets/ncoop57/rico_captions/resolve/main/captions_hierarchies_images.zip",
+    "screenshots_captions_filtered": "https://huggingface.co/datasets/ncoop57/rico_captions/resolve/main/captions_hierarchies_images_filtered.zip",
 }
 
 
@@ -77,20 +78,24 @@ class RicoDataset(datasets.GeneratorBasedBuilder):
             version=VERSION,
             description="Contains 66k+ unique UI screens. For each UI, we present a screenshot (JPG file) and the text shown on the screen that was extracted using an OCR model.",
         ),
+        datasets.BuilderConfig(
+            name="screenshots_captions_filtered",
+            version=VERSION,
+            description="Contains 25k unique UI screens. For each UI, we present a screenshot (JPG file) and the text shown on the screen that was extracted using an OCR model. Filtering was done as discussed in this paper: https://aclanthology.org/2020.acl-main.729.pdf",
+        ),
     ]
 
-    DEFAULT_CONFIG_NAME = "screenshots_captions"
+    DEFAULT_CONFIG_NAME = "screenshots_captions_filtered"
 
     def _info(self):
-        if self.config.name == "screenshots_captions":
-            features = datasets.Features(
-                {
-                    "screenshot_path": datasets.Value("string"),
-                    "caption": datasets.Value("string"),
-                    # This is a JSON obj, but will be coded as a string
-                    "hierarchy": datasets.Value("string"),
-                }
-            )
+        features = datasets.Features(
+            {
+                "screenshot_path": datasets.Value("string"),
+                "caption": datasets.Value("string"),
+                # This is a JSON obj, but will be coded as a string
+                "hierarchy": datasets.Value("string"),
+            }
+        )
 
         return datasets.DatasetInfo(
             description=_DESCRIPTION,