leetcode-notebook's Introduction

Leetcode-Notebook

Since i'm becoming a junior student in the school of CS, the ability of coding is extremely important. From today(2020-8-6), i will continuely update my notes of Leetcode till my graduation.Hope i won't become an idiot of algorithm in the future.

Mark

1 Leetcode学习路径

2 大佬Carl开路刷题

感悟：

现在还处于学习的阶段，多看看别人的解法，多总结不同的工具，不要浪费时间死磕

不得不说，自己完全做出来/看着题解做出来/过一阵子再做一遍是完全不一样的效果。想真正掌握一道不会的题目，得自己徒手做个几遍才行，不然临场发挥会遇到细节问题。

记录：

【8.27】终于刷到了20题，为自己买了一个心心念念的机械键盘。灯光巨好看，用起来，一个字，爽！

leetcode-notebook's People

Contributors

Stargazers

Watchers

leetcode-notebook's Issues

code

def train(self, X_train: np.ndarray, y_train: np.ndarray):
    """Train the classifier.

    Use the perceptron update rule as introduced in the Lecture.

    Parameters:
        X_train: a number array of shape (N, D) containing training data;
            N examples with D dimensions
        y_train: a numpy array of shape (N,) containing training labels
    """
    # TODO: implement me
    self.w = np.random.random((self.n_class, X_train.shape[1]))

    for epoch in range(self.epochs):
        if epoch % 2 == 0:
            self.lr /= 3
        for index in range(len(X_train)):
            label = y_train[index]
            data = X_train[index, :]
            w_yi = self.w[label, :]
            wyi_xi =  np.dot(w_yi, data.T)
            for class_index in range(self.n_class):
                if (class_index != label):
                    w_c = self.w[class_index, :]
                    wc_xi = np.dot(w_c, data.T)
                    if wc_xi > wyi_xi:
                        self.w[label,:] = self.w[label,:] + self.lr * data
                        self.w[class_index, :] = self.w[class_index, :] - self.lr *data
    
    # for epoch in range(self.epochs):
    #     pred = np.dot(self.w, X_train.T)
    #     for index in range(len(X_train)):
    #         pred_label= np.argmax(pred[:, index])
    #         train_label = y_train[index]
    #         if pred_label!=train_label:
    #             self.w[:, :] -=  self.lr * X_train[index, :]
    #             self.w[train_label, :] +=  2* self.lr * X_train[index, :]
    pass

def predict(self, X_test: np.ndarray) -> np.ndarray:
    """Use the trained weights to predict labels for test data points.

    Parameters:
        X_test: a numpy array of shape (N, D) containing testing data;
            N examples with D dimensions

    Returns:
        predicted labels for the data in X_test; a 1-dimensional array of
            length N, where each element is an integer giving the predicted
            class.
    """
    # TODO: implement me
    result = []

    for index in range(len(X_test)):
        data = X_test[index,:]
        prob = np.dot(self.w, data.T)
        pred = np.argmax(prob)
        result.append(pred)
    return np.array(result)

Recommend Projects