Pytorch Chain-Rules
摘要：
Derivative Rules $$ \begin{aligned} &\frac{\delta E}{\delta w^1_{jk}}=\frac{\delta E}{\delta O_k^1}\frac{\delta O_k^1}{\delta w^1_{jk}}=\frac{\delta E 阅读全文
Pytorch 感知机
摘要：
单层感知机 $$ \begin{aligned} & y = XW + b \ & y = \sum x_i*w_i+b\ \end{aligned} $$ Derivative $$ \begin{aligned} &E=\frac{1}{2}(O^1_0-t)^2\ &\frac{\delta 阅读全文
Pytorch Cross Entropy
摘要：
Entropy Uncetainly measure of surprise higher entropy = less info $$Entropy = -\sum_i P(i)\log P(i)$$ Lottery import torch a = torch.full([4], 1/4.) a 阅读全文
Pytorch GPU加速
摘要：
import torch import torch.nn.functional as F import torch import torch.nn as nn import torch.nn.functional as F import torch.optim as optim from torch 阅读全文
