Shengbin's Studio.

In this post, K means algorithm, Kernel K means and Hierarchical Clustering were implemented by Python. 1234import numpy as npfrom matplotlib import cmimport matplotlib.pyplot as plt%matplotlib inline Data GeneratorGaussian Data Generator12345678910def generateGaussianData(): nPerClass = 200 ...

In this post, a SVM classifier is implemented. The SVM is implemented with “Hard Margin” and “Soft Margin”. The “Hard Margin” is used to classify separable data, while the soft margin is used to classifier inseparable data. In addition, kernel can be used in this SVM classifier. In this post, th...

In this example, I will train a deep neural network to sort an array of 5 data. Dara Generation1import numpy as np data generation functionGenerate data set, the input of training data is arrays of integer, the length is set to 5 for this particular example. The label of a training data is 5 ind...

Image Segmentation1234import cv2import matplotlib.pyplot as pltimport matplotlib.image as mpimgimport numpy as np 1234def showImage(image): plt.figure(figsize=(8,8)) plt.imshow(image) plt.show() 12img = cv2.imread('IMG_0377_o.JPG', cv2.IMREAD_COLOR)showImage(img) Use the Canny Algorithm...

link to this project in githublink to webapplication

IntroductionWord embedding is a mapping from words to vector representations of the words. The method that represents a word as a vector or array of numbers related in someway to counts is called vector semantics. Ideally, the geometry of the vectors will capture the semantic and syntactic meanin...

IntroductionIn this post, we will train a Hidden Markov Model for part of speech tagging. The first 10K tagged sentence of ‘news’ in the brown corpus will be used to train the Hidden Markov Model. A sentence from ‘alice’ will be infered by this Hidden Markov Model. Extract data123456789101112imp...

IntroductionIn the note, a classifier of SVM(Support Vector Machine) and a classifier of CNN(Convolutional Neural Network) using transfer learning will be explored to experiment the accuracy of automatically classifying the image set of cats and dogs. Method OneFor the first method, an SVM clas...

link to this project in github