## Few Shot Learning – Siamese Network

https://miro.medium.com/max/1200/0*AuZLKEVsQpsTD3Dt Original Source Here Few Shot Learning – Siamese Network 透過few shot learning 來打造生物識別(人臉、聲紋、手寫…)模型! Photo by Green Chameleon on Unsplash 由於疫情，近期都很少出門宅在家，就來記錄一下最近有用到的技術吧~ 一般的分類問題往往都是屬於類別不多且每個類別資料量很多任務，比如MNIST 手寫資料集或 imagenet 的影像分類問題。但在生物識別的task 上我們往往沒辦法收集到那麼多資料，比如說我要建立一個人臉識別的模型，我應該不太可能跟每個我要識別的人都收集大量的照片，況且世界上的人那麽多，我也不大可能收集所有人的照片來建立分類模型。 Few shot learning 算是 meta-learning 的其中一塊，核心概念是讓模型學會學習(learn to learn)。這樣說有點懸，我們可以把它理解成: few shot learning 是要讓模型學會區分事物的差異。一個學會區分事物差異的模型，我們可以把它用在訓練集從未見過的新類別，並且可以只透過很少的樣本(few shot) 就學會區別此事物。 Siamese Network Siamese 這個詞是孿生、連體嬰的意思，表示兩個人身體相連且共享部分的器官。而siamese network 是只有兩個架構權重都相同的類神經網路組合在一起(如下右圖) 可以看到這個網路的input 是一個image pairs，而我們的目標是要訓練一個能夠區分事物差異的網路。想必聰明的你已經想到要如何使用這個網路結構了！ 首先我們要準備很多positive samples 以及 negative samples，分別表示相同類別的 imageContinue reading “Few Shot Learning – Siamese Network”

## Paper Walkthrough — Matrix Calculus for Deep Learning Part 1

Original Source Here Affine Transformation Think about the equation for a straight line, with the independent variable y and the dependent variable x: Eq.1 : Straight Line The output of a single neuron in a neural network is given by something similar to the above Eq.1: Eq.2 : Output of a single neuron in aContinue reading “Paper Walkthrough — Matrix Calculus for Deep Learning Part 1”

## Recommender system which favors newer things

Original Source Here Recommender system which favors newer things I this article I am making a proceeding on recommender system. In the recommender system time is now being included, which wasn’t in my last article “Developing a recommender system”. Now, items which were bought more recently, are more frequently recommended. It should be mentioned thatContinue reading “Recommender system which favors newer things”

## Tasks To Automate With Python

https://miro.medium.com/max/1000/0*C2wlrq7NwbWsA44M Original Source Here pip install mac-say Then create a python file to perform the task. import sysimport mac_saymac_say.say([“-f”, sys.argv[1], “-v”, “Alex”]) Then just point to a file of your choosing on the command line. python audiobook.py fileofyourchoice.txt It is typically a fast thing to check the weather, but it may also be a littleContinue reading “Tasks To Automate With Python”

## Keras vs. TensorFlow

Original Source Here What are the differences between the two? Neural Networks Basics: Weighting Functions Training Neural Networks With Keras Keras vs. TensorFlow (this article) Now that we have seen an introduction of Keras, let us look at another popular machine learning library called TensorFlow and compare the two libraries. Tensorflow is a utility createdContinue reading “Keras vs. TensorFlow”

## 3 easy machine learning projects

Original Source Here 3 easy machine learning projects Welcome back! Machine learning is increasingly becoming one of the most important industries in the world right now, so let’s talk about some of the easiest machine learning projects you can build out fairly easily. Quick note: machine learning is not easy, you will still need toContinue reading “3 easy machine learning projects”

## Data science career: is it still lucrative in 2021?

Original Source Here Data science career: is it still lucrative in 2021? Current status and prospects of the data science job market Image source: Create by Pressfoto @ Freepik The Data science job market has been booming in the last decade. Data scientists were one of the most in-demand technical experts in the IT industry.Continue reading “Data science career: is it still lucrative in 2021?”

## IBM

Original Source Here IBM Why become an AI engineer? The current and future demand is staggering. The New York Times reports candidate shortage for certified AI Engineers, with fewer than 10,000 qualified people in the world to fill these jobs, which according to earn an average salary of \$172,000 per year in the U.S. (orContinue reading “IBM”

## 5 Examples to Explain SQL Versions of Pandas Operations

Original Source Here Example 5 The last example is about one of the most frequently used Pandas functions, the groupby. We use it to compare the distinct values or categories in a column based on the values in a different column. SQL provides the group by statement for this task. Just like with Pandas, weContinue reading “5 Examples to Explain SQL Versions of Pandas Operations”