gdtop的个人笔记

important concept:1.1neural network1.2  supervised learning with neural network 1.3 structured data(num)/unstructured data(image/vedio/text)1.4traditional method(svm/knn...) standard n

 Logistic regression can be regarded as a small neural network.This part is very easy. And this is a review of logistic regression.If you have any other question, you can first FINISH machine le...

2.2 Python and Vectorization2.2.1 Vectorization We use a loop to iterate the training data, it costs long when meets a large amount of data.Vectorization is the secret why numpy have pri

This blog helps me review the course on coursera.Wrong answer:3.reshape to a column vector9."*" means the elementwise product ".dot" means matrix multiplication operation1. Question 1

I do not write any notes on my codes. Before "read" these blocks, I hope you fully understand how to build neural networks, how to train the data and how to update our parameters.Here are four program...

recommend an app named "Grammarly"3.1Neural Network Overviewfor iteration in (ntimes)z = np.dot(w.T,x)+ba = 1/(1 + np.exp(-z))dz = a - ydw = 1/m * np.dot(x,dz.T)db = 1/m * np.sum(dz)update

Deep learning is an experiment base on hyperparameters.I strongly encourage you to find a paper to write down forward and backward propagate.You need to review how to compute the derivative of m

1.1 Setting up your Machine Learning ApplicationTrain/Dev/Test setsIt depends on the data. (98/1/1)worst:Train set error: 50%(high bias) underfittingDev set error: 50%(high variance) overfit

summary:<ADAM: A METHOD FOR STOCHASTIC OPTIMIZATION>2.1 Mini-batch gradientthe size of batch: m    BGD: too long for each iterationthe size of batch:1     SGD: lose speed up(vectorizat...

The video for this week is easy.3.1  hyperparameter tuningtry randomselect a zone: log scalemodel/parallel model: depend on the computation power3.2 Batch Normalizationnormalize the hidd

Welcome to the third course which I think is the most important helpful videos in this data specialization!Week2 ML Strategy(2)2.1 Carrying out error analysisanalysis the false example and find t...

Welcome to the third course which I think is the most important helpful videos in this data specialization!Week1 ML Strategy(1)1.1 IntroductionPrepare carefully for our projects.training set

Convolutional Neural Networks1.1 Computer Visionwhy to learn cv network for large images1.2 Edge Detection Examplepy: conv_forwardtensorflow: tf.nn.conv2dkeras: Con2DHere are another convolutional com...

2.1 Case studiesLeNet-5AlexNetVGG-16 ResNets: train much deeper networkResidual blockwhy ResNet work so well? 1X1 convolution network: shrink the numbers od channels...

3.1 Object DetectionFirst, detect the goal whether it's in the image. (yc)Second, locate it.(dx,dy,dw,dh)Third, detect what object it is.(c: class)3.2 Landmark Detectionhand-made feature...

Special applications: Face recognition & Neural style transfer4.1 face recognitionemmmmmm.....one shot learningsiamese networkTriplet Lossface verification and binary classif...

1.1 sequence models1.2 notationone-hot1.3  Recurrent Neural Network Modelforward propagation1.4 backpropagation through time1.5 different types of RNNs1.6  Language model and...

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 Networ...

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. Leadin...

Improvise a Jazz Solo with an LSTM NetworkWelcome 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 even be ...

Operations on word vectorsWelcome to your first assignment of this week!Because word embeddings are very computionally expensive to train, most ML practitioners will load a pre-trained set of embeddin...

This assignment have some error with emoji.Emojify!Welcome to the second assignment of Week 2. You are going to use word vector representations to build an Emojifier.Have you ever wanted to make your ...

3.1 Various sequence to sequence architectures3.1.1 Basic Models3.1.2  Picking the most likely sentenceconditional probabilitypick most likely sentenceGreedy search(not useful)3.1.3 Beam Searchexample...

Neural Machine TranslationWelcome to your first programming assignment for this week!You will build a Neural Machine Translation (NMT) model to translate human readable dates ("25th of June, 2009") in...

Trigger Word DetectionWelcome to the final programming assignment of this specialization!In this week's videos, you learned about applying deep learning to speech recognition. In this assignment, you ...