Datasets:

c-i-ber
/

Nova

Tasks:

Object Detection

Text Generation

Zero-Shot Classification

Modalities:

Formats:

Sub-tasks:

image-captioning

multi-class-classification

Languages:

Size:

ArXiv:

Tags:

vision-language

medical-imaging

out-of-distribution

Libraries:

Dataset card Data Studio Files Files and versions

Anonymous commited on May 22, 2025

Commit

b999ac6

·

verified ·

1 Parent(s): 1aaa7b9

Update Nova.py

Files changed (1) hide show

Nova.py +0 -1

Nova.py CHANGED Viewed

@@ -76,7 +76,6 @@ class NovaDataset(GeneratorBasedBuilder):
         for idx, row in captions.iterrows():
             filename = row["filename"]
             case_id = row["case_id"]
-            image_path = os.path.join(image_dir, filename)
             record = {
                 "image": images_map.get(filename, None),
                 "filename": filename,

         for idx, row in captions.iterrows():
             filename = row["filename"]
             case_id = row["case_id"]
             record = {
                 "image": images_map.get(filename, None),
                 "filename": filename,