- คำอธิบาย :
Kuzushiji-MNIST เป็นการแทนที่แบบดรอปอินสำหรับชุดข้อมูล MNIST (ระดับสีเทา 28x28, 70,000 ภาพ) ซึ่งให้มาในรูปแบบ MNIST ดั้งเดิมและรูปแบบ NumPy เนื่องจาก MNIST จำกัดเราไว้ที่ 10 คลาส เราจึงเลือกอักขระหนึ่งตัวเพื่อเป็นตัวแทนฮิระงะนะทั้ง 10 แถวเมื่อสร้าง Kuzushiji-MNIST
เอกสารประกอบเพิ่มเติม : สำรวจในเอกสารด้วยรหัส
ซอร์สโค้ด :
tfds.image_classification.KMNISTรุ่น :
-
3.0.1(ค่าเริ่มต้น): ไม่มีบันทึกประจำรุ่น
-
ขนาดดาวน์โหลด :
20.26 MiBขนาดชุดข้อมูล :
31.76 MiBแคชอัตโนมัติ ( เอกสาร ): ใช่
แยก :
| แยก | ตัวอย่าง |
|---|---|
'test' | 10,000 |
'train' | 60,000 |
- โครงสร้างคุณสมบัติ :
FeaturesDict({
'image': Image(shape=(28, 28, 1), dtype=uint8),
'label': ClassLabel(shape=(), dtype=int64, num_classes=10),
})
- เอกสารคุณสมบัติ :
| คุณสมบัติ | ระดับ | รูปร่าง | ประเภทD | คำอธิบาย |
|---|---|---|---|---|
| คุณสมบัติDict | ||||
| ภาพ | ภาพ | (28, 28, 1) | uint8 | |
| ฉลาก | ClassLabel | int64 |
คีย์ภายใต้การดูแล (ดู
as_superviseddoc ):('image', 'label')รูป ( tfds.show_examples ):

- ตัวอย่าง ( tfds.as_dataframe ):
- การอ้างอิง :
@online{clanuwat2018deep,
author = {Tarin Clanuwat and Mikel Bober-Irizar and Asanobu Kitamoto and Alex Lamb and Kazuaki Yamamoto and David Ha},
title = {Deep Learning for Classical Japanese Literature},
date = {2018-12-03},
year = {2018},
eprintclass = {cs.CV},
eprinttype = {arXiv},
eprint = {cs.CV/1812.01718},
}