![](https://img-blog.csdnimg.cn/20201014180756916.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
深度学习
冬日and暖阳
Github: https://github.com/pengfeidip
邮箱:pengfeidip@qq.com
展开
-
PP-LCNet-A Lightweight CPU Convolutional Neural Network
百度PP-LCNET论文阅读原创 2021-12-08 11:02:07 · 850 阅读 · 0 评论 -
nvidia-fp16-fp32
https://blog.csdn.net/qq_36533552/article/details/105885714原创 2021-04-07 13:12:32 · 380 阅读 · 0 评论 -
BatchNormalization-归一化实现
batch-normalization的维度是按照mini-batch的维度进行的。也就是对于维度为[n, c ,h , w]的input来说,其对应的BN的λ参数的维度是 [1, c, 1, 1]import torchimport torch.nn as nninput = torch.rand(4, 3, 5, 5, dtype=torch.float32)print(input[0, 0, 0, :]) # 输入# 自己写的batch-normalization 减均值,除以方差原创 2021-02-25 19:09:14 · 778 阅读 · 0 评论 -
DenseNet网络结构可视化_prototxt
网址:http://ethereon.github.io/netscope/#/editor# Enter your network definition here.# Use Shift+Enter to update the visualization. layer { name: "Data1" type: "ImageData" top: "Data1" top: "label" include { phase: TRAIN } transform_p原创 2020-11-23 17:35:30 · 295 阅读 · 0 评论