small patch for answer variants

DUDE_imdb_loader.py

@@ -36,7 +36,7 @@ from datasets import load_dataset_builder, load_dataset, logging
 
 logger = logging.get_logger(__name__)
 
-PIL_Image.MAX_IMAGE_PIXELS = None #933120000
+PIL_Image.MAX_IMAGE_PIXELS = None  # 933120000
 
 MAX_PAGES = 50
 MAX_PDF_SIZE = 100000000  # almost 100MB
@@ -63,7 +63,9 @@ def get_images_pdf2image(document_filepath, chunksize=10):
         try:
             images.extend(
                 pdf2image.convert_from_path(
-                    document_filepath, first_page=page, last_page=min(page + chunksize - 1, maxPages)
+                    document_filepath,
+                    first_page=page,
+                    last_page=min(page + chunksize - 1, maxPages),
                 )
             )
         except Exception as e:
@@ -78,7 +80,9 @@ def pdf_to_images(document_filepath, converter="PyPDF2"):
             page_image_name = document_filepath.replace("PDF", "images").replace(
                 ".pdf", f"_{page_idx}.jpg"
             )
-            page_image_names.append(page_image_name.replace(page_image_dir, page_image_dir.split('/')[-1]))  # without dir
+            page_image_names.append(
+                page_image_name.replace(page_image_dir, page_image_dir.split("/")[-1])
+            )  # without dir
             if not os.path.exists(page_image_name):
                 page_image.convert("RGB").save(page_image_name)
         return page_image_names
@@ -211,7 +215,7 @@ def get_document_info(documents_metadata, docId):
 
 
 def format_answers(answers_list):
-    answers_list = list(set([answer.lower() for answer in answers_list]))
+    answers_list = list(set([answer for answer in answers_list]))  # standardization is done
     return answers_list
 
 
@@ -220,7 +224,6 @@ def create_imdb_record_from_json(
 ):
 
     docId = record["docId"].split("_")[0]
-    # document_filepath = documents_dict[docId]
     try:
         num_pages, page_image_names = get_document_info(documents_metadata, docId)
         document_ocr_info = documents_ocr_info[docId]
@@ -237,12 +240,13 @@ def create_imdb_record_from_json(
     else:
         answers = None
 
-    if include_variants:
-        import pdb; pdb.set_trace()  # breakpoint 03e74e0e //
-        
+    if include_variants and record['answers_variants']:
+        answers += record["answers_variants"]
 
-    #fix if wrongly saved pagenames
-    page_image_names = [split+image_name for image_name in page_image_names if image_name.startswith('/')]
+    # fix if wrongly saved pagenames
+    # page_image_names = [
+    #     split + image_name for image_name in page_image_names if image_name.startswith("/")
+    # ]
 
     imdb_record = {
         "question_id": record["questionId"],
@@ -253,7 +257,7 @@ def create_imdb_record_from_json(
         "ocr_tokens": document_ocr_info["ocr_tokens"],
         "ocr_normalized_boxes": document_ocr_info["ocr_boxes"],
         "set_name": split,
-        "answers": answers,
+        "answers": answers,  # list
         "answer_page": None,
         "extra": {
             # 'question_type': record['qtype'],
@@ -267,14 +271,20 @@ def create_imdb_record_from_json(
 
 
 def create_imdb_from_json(
-    data, documents_metadata, documents_ocr_info, split, version, include_answers=True
+    data,
+    documents_metadata,
+    documents_ocr_info,
+    split,
+    version,
+    include_answers=True,
+    include_variants=False,
 ):
     imdb_header = create_header(split, version, include_answers)
 
     imdb_records = []
     for record in tqdm(data):
         imdb_record = create_imdb_record_from_json(
-            record, documents_metadata, documents_ocr_info, split, include_answers
+            record, documents_metadata, documents_ocr_info, split, include_answers, include_variants
         )
         if imdb_record:
             imdb_records.append(imdb_record)
@@ -284,12 +294,53 @@ def create_imdb_from_json(
     return imdb
 
 
-if __name__ == "__main__":
-    dataset = load_dataset(
-        "../DUDE_loader/DUDE_loader.py",
-        "DUDE",
-        data_dir="/home/jordy/Downloads/DUDE_train-val-test_binaries",
+def parse_arguments():
+    import argparse
+
+    parser = argparse.ArgumentParser(
+        description="Instantiate HuggingFace dataloader and convert to ImDB format"
+    )
+
+    parser.add_argument(
+        "--redo-imdb-build",
+        action="store_true",
+        default=False,
+        help="Whether to rebuild the imdb record and overwrite",
+    )
+
+    # Optional
+    parser.add_argument(
+        "--no-include-variants",
+        action="store_false",
+        default=True,
+        help="Whether to include variants as full ground truths",
+    )
+
+    # Overwrite/Pass HF parameters
+    parser.add_argument(
+        "--DUDE_config",
+        type=str,
+        default="Amazon_original",
+        help="HF Config to load to control OCR version",
+    )
+    parser.add_argument(
+        "--data_dir",
+        type=str,
+        default="/home/jordy/Downloads/DUDE_train-val-test_binaries",
+        help="Load PDFs and store images",
     )
+
+    return parser.parse_args()
+
+
+if __name__ == "__main__":
+    """
+    Parse and redo
+    """
+    args = parse_arguments()
+
+    dataset = load_dataset("jordyvl/DUDE_loader", args.DUDE_config, data_dir=args.data_dir)
+
     splits = dataset.keys()
 
     for split in splits:
@@ -297,7 +348,7 @@ if __name__ == "__main__":
         OCR_paths = []
         document_paths = []
         for i, x in enumerate(dataset[split]):
-            if x["data_split"] != split: #redundant check
+            if x["data_split"] != split:  # redundant check
                 continue
             if x["document"] not in document_paths:
                 document_paths.append(x["document"])
@@ -371,12 +422,10 @@ if __name__ == "__main__":
             save_json(documents_ocr_filename, documents_ocr_info)
 
         imdb_filename = f"imdb_{split}.npy"
-        if os.path.exists(imdb_filename):
+        if os.path.exists(imdb_filename) and not args.redo_imdb_build:
             print(f"Loading from disk: {imdb_filename}")
             imdb = np.load(imdb_filename, allow_pickle=True)
 
-            import pdb; pdb.set_trace()  # breakpoint ff49174b //
-
         else:
             imdb = create_imdb_from_json(
                 dataset[split],  # .select(split_indices),
@@ -385,6 +434,6 @@ if __name__ == "__main__":
                 split=split,
                 version="0.1",
                 include_answers=True,
+                include_variants=(not args.no_include_variants),
             )
-            np.save(imdb_filename, imdb)
-            #Missing:  'a7b500f3e0244a50571769aaef4fabc7' a7b500f3e0244a50571769aaef4fabc7
+            np.save(imdb_filename, imdb)

DUDE_loader.py

@@ -49,9 +49,8 @@ _URLS = {
     # "binaries": "https://huggingface.co/datasets/jordyvl/DUDE_loader/resolve/main/data/DUDE_binaries.tar.gz", #
     # "annotations": "https://huggingface.co/datasets/jordyvl/DUDE_loader/resolve/main/data/DUDE_dataset-sample_gt.json" #"
     "binaries": "https://huggingface.co/datasets/jordyvl/DUDE_loader/resolve/main/data/DUDE_train-val-test_binaries.tar.gz",  # DUDE_train-val-test_binaries.tar.gz
-    "annotations": "https://zenodo.org/record/7652563/files/DUDE_gt_release-candidate_trainval.json?download=1",
+    "annotations": "https://zenodo.org/record/7662385/files/DUDE_gt_release-candidate_trainval.json?download=1",
 }
-
 SKIP_DOC_IDS = ["ef03364aa27a0987c9870472e312aceb", "5c5a5880e6a73b4be2315d506ab0b15b"]
 
 

test_loader.py

@@ -22,7 +22,7 @@ from codetiming import Timer
 for binding in ["dict_annotations (new)"]: #"dict_PDF", 
     with Timer(name=f"{binding}", text=binding + " Elapsed time: {:.4f} seconds"):
         if binding == "dict_annotations (new)":
-            ds = load_dataset("../DUDE_loader/DUDE_loader.py", 'Amazon_original', data_dir="/home/jordy/Downloads/DUDE_train-val-test_binaries", ocr_engine='Azure') #ignore_verifications=True, , writer_batch_size=10
+            ds = load_dataset("jordyvl/DUDE_loader", 'Amazon_original', data_dir="/home/jordy/Downloads/DUDE_train-val-test_binaries") #ignore_verifications=True, , writer_batch_size=10
         else:
             ds = load_dataset("jordyvl/DUDE_loader", revision='db20bbf751b14e14e8143170bc201948ef5ac83c')