downsampled_imagenet

ชุดข้อมูลที่มีภาพ 2 ความละเอียด (ดูชื่อการกำหนดค่าสำหรับข้อมูลเกี่ยวกับความละเอียด) ใช้สำหรับการประมาณค่าความหนาแน่นและการทดลองการสร้างแบบจำลองเชิงกำเนิด

สำหรับ ImageNet ที่ปรับขนาดสำหรับการเรียนรู้แบบมีผู้สอน ( ลิงก์ ) โปรดดู imagenet_resized

แยก ตัวอย่าง
'train' 1,281,149
'validation' 49,999
  • โครงสร้างคุณลักษณะ :
FeaturesDict({
    'image': Image(shape=(None, None, 3), dtype=uint8),
})
  • เอกสารคุณสมบัติ :
ลักษณะเฉพาะ ระดับ รูปร่าง Dประเภท คำอธิบาย
คุณสมบัติDict
ภาพ ภาพ (ไม่มี, ไม่มี, 3) uint8
  • คีย์ภายใต้การดูแล (ดู as_supervised doc ): None

  • การอ้างอิง :

@article{DBLP:journals/corr/OordKK16,
  author    = {A{"{a} }ron van den Oord and
               Nal Kalchbrenner and
               Koray Kavukcuoglu},
  title     = {Pixel Recurrent Neural Networks},
  journal   = {CoRR},
  volume    = {abs/1601.06759},
  year      = {2016},
  url       = {http://arxiv.org/abs/1601.06759},
  archivePrefix = {arXiv},
  eprint    = {1601.06759},
  timestamp = {Mon, 13 Aug 2018 16:46:29 +0200},
  biburl    = {https://dblp.org/rec/bib/journals/corr/OordKK16},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}

downsampled_imagenet/32x32 (ค่าเริ่มต้น)

  • คำอธิบาย การกำหนดค่า : ชุดข้อมูลที่ประกอบด้วยรูปภาพ Train และ Validation ที่มีความละเอียด 32x32

  • ขนาดการดาวน์โหลด : 3.98 GiB

  • ขนาดชุดข้อมูล : 3.05 GiB

  • รูป ( tfds.show_examples ):

การสร้างภาพ

downsampled_imagenet/64x64

  • คำอธิบาย การกำหนดค่า : ชุดข้อมูลที่ประกอบด้วยรูปภาพ Train และการตรวจสอบความถูกต้องของความละเอียด 64x64

  • ขนาดการดาวน์โหลด : 11.73 GiB

  • ขนาดชุดข้อมูล : 10.80 GiB

  • รูป ( tfds.show_examples ):

การสร้างภาพ