cs231n笔记（5）（草稿）

1. SVM 代码解析

损失函数的计算公式如下：
$L(W)=\frac{1}{N}{\sum }_{i=1}^N{L}_i(f(x_i,W),y_i)+\lambda R(W)$

而损失函数关于权重$W$的求导公式如下：
$$\begin{gathered} {\nabla }_{W}L(W)=\frac{1}{N}{\sum }_{i=1}^N{\nabla }_W{L}_i(x_i,y_i,W)+\lambda {\nabla }_{W}R(W) \\ =\frac{1}{N}{\sum }_{i=1}^N{x}_i\ast y_i+W \end{gathered}$$

对于Hinge Loss， $L_i(f(x_i,W),y_i)={\sum }_{j\ne y_i}max(0,(W\ast x_i{)}_j-(W\ast x_i{)}_{y_i}+\Delta )$

在实验中我们取 $\Delta =1$, 正则化项 $\lambda R(W)=\frac{1}{2}||W|{|}_F^2$, 预测结果 $scor{e}_i=W\ast x_i\in R^c$，c为数据的类别数。

代码中的margin就是第 $i$ 个样本在第 $j$ 类的loss，一个内循环结束后就得到该样本的损失值。

Hinge loss 导数:
$\frac{\partial L}{\partial W}=$

Softmax Loss 导数：

def svm_loss_naive(W, X, y, reg):
  """
  Structured SVM loss function, naive implementation (with loops).

  Inputs have dimension D, there are C classes, and we operate on minibatches
  of N examples.

  Inputs:
  - W: A numpy array of shape (D, C) containing weights.
  - X: A numpy array of shape (N, D) containing a minibatch of data.
  - y: A numpy array of shape (N,) containing training labels; y[i] = c means
    that X[i] has label c, where 0 <= c < C.
  - reg: (float) regularization strength

  Returns a tuple of:
  - loss as single float
  - gradient with respect to weights W; an array of same shape as W
  """
  dW = np.zeros(W.shape) # initialize the gradient as zero

  # compute the loss and the gradient
  num_classes = W.shape[1]
  num_train = X.shape[0]
  loss = 0.0
  for i in range(num_train):
    scores = X[i].dot(W)
    correct_class_score = scores[y[i]]
    for j in range(num_classes):
      if j == y[i]:
        continue
      margin = scores[j] - correct_class_score + 1 # note delta = 1
      if margin > 0:
        loss += margin

  # Right now the loss is a sum over all training examples, but we want it
  # to be an average instead so we divide by num_train.
  loss /= num_train

  # Add regularization to the loss.
  loss += reg * np.sum(W * W)

  
  dW = (W - h)/h
  
  
  return loss, dW

2. Hinge loss ,softmax loss梯度计算

3. Svm ,softmax训练的损失函数及测试结果 权重可视化结果