Semantic Scholar
Open Access
2016
1669 sitasi
A guide to convolution arithmetic for deep learning
Vincent Dumoulin
Francesco Visin
Abstrak
We introduce a guide to help deep learning practitioners understand and manipulate convolutional neural network architectures. The guide clarifies the relationship between various properties (input shape, kernel shape, zero padding, strides and output shape) of convolutional, pooling and transposed convolutional layers, as well as the relationship between convolutional and transposed convolutional layers. Relationships are derived for various cases, and are illustrated in order to make them intuitive.
Topik & Kata Kunci
Penulis (2)
V
Vincent Dumoulin
F
Francesco Visin
Akses Cepat
PDF tidak tersedia langsung
Cek di sumber asli →Informasi Jurnal
- Tahun Terbit
- 2016
- Bahasa
- en
- Total Sitasi
- 1669×
- Sumber Database
- Semantic Scholar
- Akses
- Open Access ✓