Hansry的博客

1. 简介SoftmaxSoftmaxSoftmax 函数SoftmaxSoftmaxSoftmax 函数在神经网络分类是十分常用的函数，如下所示，在神经元output layer中，可以输出一个R4R4R^{4} 维度的向量，来进行分类，例如输出层为向量 O=[0.2,0.1,0.4,0.3]O=[0.2,0.1,0.4,0.3]O=[ 0.2,0.1,0.4,0.3 ], 可根据向量中...

Improvise a Jazz Solo with an LSTM Network​ Welcome to your final programming assignment of this week! In this notebook, you will implement a model that uses an LSTM to generate music. You will eve...

Character level language model - Dinosaurus landWelcome to Dinosaurus Island! 65 million years ago, dinosaurs existed, and in this assignment they are back. You are in charge of a special task. Lead...

Building your Recurrent Neural Network - Step by StepWelcome to Course 5’s first assignment! In this assignment, you will implement your first Recurrent Neural Network in numpy.Recurrent Neural Ne...

1.词汇表征深度学习已经给自然语言处理（Natural Language Process， NLP）带来革命性的变革。其中一个很关键的概念是词嵌入（word embedding），这是语言表示的一种方式，可以让算法自动的了解一些类似的词，例如男人对女人（man to woman），国王对王后（king to queen）等。在此前的表示中，我们用的一直都是词典的编号来表达一个词，例如Man...

一. 序列模型常用领域如下：Speech recognition（语音识别） Music generation (音乐生成): In this cases，only the Y is a sequence, and X can be a single integer or the empty set. Y Sentiment classification (情感分析): X is a ...

Autonomous driving - Car detectionWelcome to your week 3 programming assignment. You will learn about object detection using the very powerful YOLO model. Many of the ideas in this notebook are describ

Welcome to the first assignment of week 2. In this assignment, you will: 1. Learn to use Keras, a high-level neural networks API (programming framework), written in Python and capable of running on to

Convolutional Neural Networks: Step by StepWelcome to Course 4’s first assignment! In this assignment, you will implement convolutional (CONV) and pooling (POOL) layers in numpy, including both forward

TensorFlow TutorialWelcome to this week’s programming assignment. Until now, you’ve always used numpy to build neural networks. Now we will step you through a deep learning framework that will allow yo

Optimization MethodsUntil now, you’ve always used Gradient Descent to update the parameters and minimize the cost. In this notebook, you will learn more advanced optimization methods that can speed up

import numpy as npfrom testCases import *from gc_utils import sigmoid,relu,dictionary_to_vector,vector_to_dictionary,gradients_to_vectorfrom testCases import gradient_check_n_test_caseGradient_check

如果数据集没有很大，同时在训练集上又拟合得很好，但是在测试集的效果却不是很好，这时候就要使用正则化来使得其拟合能力不会那么强。import numpy as npimport sklearnimport matplotlib.pyplot as pltimport sklearn.datasetsfrom reg_utils import load_2D_dataset,compute_co

import numpy as npimport matplotlib.pyplot as pltimport sklearnimport sklearn.datasetsfrom init_utils import sigmoid, relu, compute_loss, forward_propagation, backward_propagationfrom init_utils i

1.调试处理关于训练神经网络，要处理的参数的数量，从学习速率αα\alpha 到momentum ββ\beta 同时还有Adam优化算法的参数 ββ\beta ，ββ\beta and εε\varepsilon，也p许还需要选择层数 layer_size, 不同层中隐藏单元的数量 layer_units ,有可能还需要选择学习率衰减 learning_rate_decay,还有需要选...

1.Mini-batch梯度下降法对于m个数据样本，在训练神经网络的时候，最常用的是向量化，如：X=[x(1),x(2),...,x(m)],x(nx,m)X=[x(1),x(2),...,x(m)],x(nx,m)X=[x^{(1)},x^{(2)},...,x^{(m)}],x(n_{x},m)（维数） ，但是当样本足够大的时候，例如5w，虽然向量化了，在back propagation的...

1.测试集在配置训练、验证和测试数据集的过程中做出正确决策会更高效的创建神经网络在机器学习中，经常将样本分为训练集、验证集和测试集三部分。对于早期的机器学习的数据而言，经常使用 70%/30% 来划分验证集和测试集或者使用 60%/20%/20% 来划分训练集、验证集和测试集。 在大数据时代中，假如有1000000个数据，那么验证集就不需要太大，有10000条验证数据或者10000条测...

1.Packageimport numpy as np #scientific compute packageimport matplotlib.pyplot as plt #graphs packageimport h5py #contact with h5 fileimport scipy from scipy import ndimage #import our image a

1.Packages1.numpy is the main package for scientific computing with Python 2.matplotlib is a library to plot graphs in Python 3.dnn_unils provides some necessary functions for this notebook such as s

Planar data classification with one hidden layer1.常用的Python Librarynumpy：is the fundamental package for scientific computing with Python sklearn：provides simple and efficient tools for data mining and

1.主要内容：1、使用房价预测的例子来建立最简单的由神经元组成的神经网络模型；2、逐步增强神经网络模型的复杂度，建立标准的神经网络模型结构；3、从监督式学习入手，介绍了不同的神经网络类型，包括Standard NN、CNN、RNN。4、数据分为结构化数据和非结构化数据。在处理非结构化数据上，深度学习的处理能力近年来提高的非常快，例如图像处理、语音识别和自动翻译等。5、深度学习飞...