imagenet_sketch

  • תיאור :

ImageNet-Sketch מורכב מ-50,889 תמונות סקיצה בשחור-לבן, 50 עבור כל אחת מ-1000 מחלקות ImageNet. תמונות אלו נאספו במקור מחיפוש תמונות של Google עבור "סקיצה של __". 100 תמונות נאספו ולאחר מכן סוננו באופן ידני. עבור כיתות עם פחות מ-50 תמונות טובות, תמונות נוספות נבנו על ידי היפוך או סיבוב.

לְפַצֵל דוגמאות
'test' 50,889
  • מבנה תכונה :
FeaturesDict({
    'file_name': Text(shape=(), dtype=string),
    'image': Image(shape=(None, None, 3), dtype=uint8),
    'label': ClassLabel(shape=(), dtype=int64, num_classes=1000),
})
  • תיעוד תכונה :
תכונה מעמד צוּרָה Dtype תיאור
FeaturesDict
שם קובץ טֶקסט חוּט
תמונה תמונה (אין, אין, 3) uint8
תווית ClassLabel int64

רְאִיָה

  • ציטוט :
@inproceedings{wang2019learning,
        title={Learning Robust Global Representations by Penalizing Local Predictive Power},
        author={Wang, Haohan and Ge, Songwei and Lipton, Zachary and Xing, Eric P},
        booktitle={Advances in Neural Information Processing Systems},
        pages={10506--10518},
        year={2019}
}