原创 A Content Transformation Block for Image Style Transfer (CVPR2019, 風格遷移)
Kotovenko, Dmytro, et al. “A Content Transformation Block for Image Style Transfer.” CVPR 2019. 風格遷移問題: X→YX \right
2019-08-08 06:04:26
18
原创 Unsupervised pixel-level domain adaptation with generative adversarial networks (DA+ 圖像轉換)
paper: Bousmalis, Konstantinos, et al. “Unsupervised pixel-level domain adaptation with generative adversarial net
2019-08-05 21:04:16
原创 Discriminative deep metric learning for face verification in the wild 度量學習(CVPR2014)
Hu, Junlin, Jiwen Lu, and Yap-Peng Tan. “Discriminative deep metric learning for face verification in the wild.” CV
2019-08-05 21:04:16
2
原创 馬氏距離
1. 協方差矩陣是半正定矩陣 對於向量xxx,設其均值爲uuu。那麼協方差矩陣Σ=E[(x−u)(x−u)T]\Sigma=E[(x-u)(x-u)^T]Σ=E[(x−u)(x−u)T] yTΣy=yTE[(x−u)(x−u)T]
2019-08-02 21:46:52
原创 Beyond sharing weights for deep domain adaptation (PAMI 2018) ---Transfer Learning
Rozantsev, Artem, Mathieu Salzmann, and Pascal Fua. “Beyond sharing weights for deep domain adaptation.” IEEE trans
2019-07-09 22:58:27
原创 Deep Domain Confusion: Maximizing for Domain Invariance
Tzeng E , Hoffman J , Zhang N , et al. Deep Domain Confusion: Maximizing for Domain Invariance[J]. Computer Science
2019-07-08 22:25:04
13
原创 筆記： Gradient Reversal Layer (unsupervised domain adaptation by backpropagation. ICML 2015)
paper: Ganin, Yaroslav, and Victor Lempitsky. “Unsupervised domain adaptation by backpropagation.” ICML 37. JMLR. o
2019-07-07 22:02:19
208
原创 PyTorch 模型參數和optimizer
網絡參數 ( parameters()和named_parameters()) model爲網絡， 打印層結構： ret=[*model.modules()] for layer in ret: #print(type
2019-04-11 21:57:03
原创 pandas讀取csv並繪製散點圖
讀取 loc='checkpoints/mnist_vae2/mu_and_sigma.csv' import pandas as pd import numpy as np X=pd.DataFrame.from_csv(loc)
2019-03-20 21:49:50
原创 frp內網穿透實例
（目前僅實現了SSH功能） 問題： 具有公網IP的計算機A需要訪問僅有內網IP的gpu服務器C。 情況: 本來想通過具有內網和外網ip的跳板機D實現轉發。但不知D做了何種設置，A始終連不上D。 解決方案： 使用了一臺具有公網ip的雲服
2019-03-06 15:44:16
原创 原點矩與中心矩
K階原點矩: E(xk)E(x^k)E(xk),1階原點矩是數學期望。 K階中心矩: E(x−E(x))kE(x-E(x))^kE(x−E(x))k,2階中心矩是方差。
2019-02-27 22:44:41
4
原创 多元高斯分佈/多元正態分佈
這些知識，動不動就忘。爲了不白選了一門數學課，還是把他記下來。 f(x)=1(2π)n/2∣C∣12exp{(x−μ)TC−1(x−μ)}f(x)= \frac{1}{ (2\pi)^{n/2} {|C|}^{\frac{1}{2}}
2019-02-27 22:44:30
2
原创 主子式、順序主子式、餘子式、代數餘子式
K階主子式[1]^{[1]}[1] 以3階行列式爲例： [a1a2a3b1b2b3c1c2c3]\begin{bmatrix} a_1 & a_2 & a_3 \\ b_1 & b_2 & b_3\\
2019-02-23 22:46:34
2
原创 Tensorflow 教程筆記---Custom layers
import tensorflow as tf tf.enable_eager_execution() Layers: common sets of useful operations # In the tf.keras.laye
2019-02-10 22:33:05
6
原创 Tensorflow 教程筆記---Custom training basics
import tensorflow as tf tf.enable_eager_execution() Variable operator x = tf.zeros([10, 10]) x += 2 # This is equi
2019-02-08 17:41:44
88