shifted to binary, do to images map later

rvl_cdip_multi.py

@@ -4,10 +4,11 @@ from typing import List
 import datasets
 import pdf2image
 from tqdm import tqdm
+import io
 
 datasets.logging.set_verbosity_info()
 logger = datasets.logging.get_logger(__name__)
-
+MODE="binary"
 
 _CITATION = """\
 @inproceedings{jordyvlRVLmulti,
@@ -23,8 +24,6 @@ _DESCRIPTION = """\
 blabla
 """
 
-_VERSION = "1.0.0"
-
 _HOMEPAGE = "TBD"
 
 
@@ -74,10 +73,11 @@ def batched_conversion(pdf_file):
         )
     return images
 
+def open_pdf_binary(pdf_file):
+    with open(pdf_file, "rb") as f:
+        return f.read()
 
 class RvlCdipMulti(datasets.GeneratorBasedBuilder):
-    VERSION = datasets.Version("1.0.0")
-
     BUILDER_CONFIGS = [
         datasets.BuilderConfig(
             name="default",
@@ -123,7 +123,7 @@ class RvlCdipMulti(datasets.GeneratorBasedBuilder):
             description=_DESCRIPTION,
             features=datasets.Features(
                 {
-                    "file": datasets.Sequence(datasets.Image()),
+                    "file": datasets.Value("binary"), #datasets.Sequence(datasets.Image()),
                     "labels": datasets.features.ClassLabel(names=_CLASSES),
                 }
             ),
@@ -166,12 +166,14 @@ class RvlCdipMulti(datasets.GeneratorBasedBuilder):
 
         for i, path in tqdm(enumerate(Path(archive_path).glob("**/*/*")), desc=f"{archive_path}"):
             if path.suffix in extensions:
-
                 try:
-                    images = batched_conversion(path)
+                    images = open_pdf_binary(path)
+                    #images = batched_conversion(path)
                     yield path.name, {
                         "file": images,
                         "labels": labels.encode_example(path.parent.name.lower()),
                     }
                 except Exception as e:
                     logger.warning(f"{e} failed to parse {i}")
+
+# can do a map afterwards

test_loader.py

@@ -12,7 +12,7 @@ else:
 
 
 if stepwise:
-    ds = load_dataset("jordyvl/rvl_cdip_multi", cache_dir="/mnt/lerna/data/HFcache")
+    ds = load_dataset(DATASET, cache_dir="/mnt/lerna/data/HFcache")
     import pdb; pdb.set_trace()  # breakpoint 1898e7e0 //
     
     builder = load_dataset_builder(DATASET)