Author: Gallon Wong

Introduction to Augmented Reality and ARCore (Coursera)

Gallon WongLeave a comment

Start 20200718 (hurdles to overcome…)

  1. Contents (brief)
    1. WK1 – AR for nonprofits
    2. The hardware that makes mobile AR work
      1. enable motion tracking for AR
        1. accelerometer: measure acceleration which is change in velocity;
        2. gyroscope: measures and/or maintains orientation and angular velocity, and ARCore ensures that the digital assets respond correctly.
        3. phone camera supplies a live feed of the surrounding real world;
      2. enable location-based AR
        1. Magnetometer: simple orientation related to the Earth’s magnetic field, which direction is North, allowing it to auto-rotate digital maps depending on your physical orientation. key to location-based AR apps;
        2. GPS: geolocation and time information to a GPS receiver
      3. enable view of real world with AR
        1. display: crisp imagery and displaying 3D rendered assets
    3. Components of realism (AR object has to act like its equivalent in the real world)
      1. Placing and Positioning assets: a point in reality
      2. Scale and the size of assets
      3. Occlusion: an image or object is blocked by another
      4. Lighting for increased realism
      5. Solid augmented assets: AR objects should never overlap with real-world objects;
      6. Context awareness (most difficult)
    4. Tracking in AR
      1. Outside-in tracking: input from external devices, gain functionalities but lose portability;
      2. Inside-out tracking: cameras and sensors are built right into the body of the device;
    5. Motion Tracking: show convincing AR objects; Tech (Simultaneous Localization Mapping or SLAM)
    6. Environmental understanding: feature points and plane-finding
      1. Plane-finding: a term for ARCore’s ability to detect and generate flat surfaces from features point captured;
    7. Light estimation: create more believable AR apps, games and experiences;
  2. Contents (details)
  3. References
Introduction to Augmented Reality and ARCore (Coursera)

Book – NLP with Python

Gallon WongLeave a comment
  1. Objectives:
    1. prepare DMU module and project
    2. Building robust systems to perform linguistic tasks with technological applications;
    3. Using linguistic algorithms and data structures in robust language processing software;
    4. work in Alicould platform
    5. python crawls
    6. python data
    7. NLP with ChatRobots
    8. NLP with AutoCode\
    9. NLP with content generation / design…
    10. NLP with voice-recognition
    11. NLP with visual-recognition?
    12. Cloud tech: Azure / AliCloud…Amazon?…
    13. IoT + Industrial 4.0 + Cloud
  2. Book Contents (Brief)
    1. CH1 – Language Processing and Python
    2. CH2 – Accessing Text Corpora and Lexical Resources
    3. CH3 – Processing Raw Text
    4. CH4 – Writing Structured Programs
    5. CH5 – Categorizing and Tagging Words
    6. CH6 – Learning to Classify Text
    7. CH7 – Extracting Information from Text
    8. CH8 – Analyzing Sentence Structure
    9. CH9 – Building Feature-Based Grammars
    10. CH10 – Analyzing the Meaning of Sentences
    11. CH11 – Managing Linguistic Data
  3. Book Contents (Details)
    1. CH1 – Language Processing and Python
      1. Chapter Questions
        1. how can we automatically extract key words and phrases that sum up the style and content of a text?
        2. Some of the interesting challenges of  natural language processing
      2. Computing with language: Texts and Words
        1. Python interpreter
      3. Closer look at Python
    2. CH2 – Accessing Text Corpora and Lexical Resources
    3. CH3 – Processing Raw Text
    4. CH4 – Writing Structured Programs
    5. CH5 – Categorizing and Tagging Words
    6. CH6 – Learning to Classify Text
    7. CH7 – Extracting Information from Text
    8. CH8 – Analyzing Sentence Structure
    9. CH9 – Building Feature-Based Grammars
    10. CH10 – Analyzing the Meaning of Sentences
    11. CH11 – Managing Linguistic Data
  4. References
    1. Getting Started, Python for beginners. https://opentechschool.github.io/python-beginners/en/getting_started.html#what-is-python-exactly
    2. Jurafsky & Martin (2018). Speech and Language Processing. Prentice Hall.
    3. Open source library of Natural Language Toolkit (NLTK). http://www.nltk.org/
    4. http://docs.python.org/
    5. Python online tutorials. https://docs.python.org/3.8/tutorial/index.html
    6. Why Python?
      1. shallow learning curve
      2. Python allows you to type directly into the interactive interpreter
    7. NLTK 3.5 guide to install Python 3.0. https://docs.python-guide.org/starting/install3/win/#install3-windows
      1. NLTK. http://www.nltk.org/install.html
        1. Install Python 3.8 (x86) (aviod 64bits)
          1. eg. C:\Users\Username\AppData\Roaming\Python\Python38\site-packages
        2. Install Numpy (optional): https://www.scipy.org/scipylib/download.html
        3. Install NLTK: http://pypi.python.org/pypi/nltk
        4. Test installation: Start>Python38, then type import nltk
        5. Installing NLTK data. http://www.nltk.org/data.html
          1. command line type python -m nltk.downloader popular, or in the Python interpreter import nltk; nltk.download(‘popular’)
        6. Building ?Python 3.8 environments and interaction in Windows with NLTK / pip setup…
    8. Requirements
      1. Python version 3
      2. NLTK version 3.5
      3. NumPy
      4. Matplotlib
      5. NetworkX
      6. Prover9
    9. Practice platform
      1. https://notebooks.azure.com/, project NLP with Python
Book – NLP with Python

D-Biz research of HK

Gallon WongLeave a comment
  1. 核心: 中小企進行數碼轉型的需求
  2. References
    1. 【遙距營商計劃】個案解讀：用D-Biz買高價筆電公司用？重點幫助數碼轉型並非津貼買電腦. https://sme.hket.com/article/2647448/%E3%80%90%E9%81%99%E8%B7%9D%E7%87%9F%E5%95%86%E8%A8%88%E5%8A%83%E3%80%91%E5%80%8B%E6%A1%88%E8%A7%A3%E8%AE%80%EF%BC%9A%E7%94%A8D-Biz%E8%B2%B7%E9%AB%98%E5%83%B9%E7%AD%86%E9%9B%BB%E5%85%AC%E5%8F%B8%E7%94%A8%EF%BC%9F%E9%87%8D%E9%BB%9E%E5%B9%AB%E5%8A%A9%E6%95%B8%E7%A2%BC%E8%BD%89%E5%9E%8B%E4%B8%A6%E9%9D%9E%E6%B4%A5%E8%B2%BC%E8%B2%B7%E9%9B%BB%E8%85%A6
    2. 【遙距營商計劃】個案解讀：一人公司如何由零開始申請D-Biz. https://sme.hket.com/article/2648577/%E3%80%90%E9%81%99%E8%B7%9D%E7%87%9F%E5%95%86%E8%A8%88%E5%8A%83%E3%80%91%E5%80%8B%E6%A1%88%E8%A7%A3%E8%AE%80%EF%BC%9A%E4%B8%80%E4%BA%BA%E5%85%AC%E5%8F%B8%E5%A6%82%E4%BD%95%E7%94%B1%E9%9B%B6%E9%96%8B%E5%A7%8B%E7%94%B3%E8%AB%8BD-Biz
    3. D-Biz全城中伏 創科局呃盡中小企 10萬資助變$9,000. https://hk.appledaily.com/finance/20200613/3267QDIQRT5UM7DDGAK2TNQZ3U/
    4. 業界批D-Biz「全額資助」縮水 未能如期完成審批 薛永恒致歉. https://news.mingpao.com/pns/%E6%B8%AF%E8%81%9E/article/20200617/s00002/1592332727717/%E6%A5%AD%E7%95%8C%E6%89%B9d-biz%E3%80%8C%E5%85%A8%E9%A1%8D%E8%B3%87%E5%8A%A9%E3%80%8D%E7%B8%AE%E6%B0%B4-%E6%9C%AA%E8%83%BD%E5%A6%82%E6%9C%9F%E5%AE%8C%E6%88%90%E5%AF%A9%E6%89%B9-%E8%96%9B%E6%B0%B8%E6%81%92%E8%87%B4%E6%AD%89
D-Biz research of HK

NLP

Gallon WongLeave a comment

 

  1. L1 – Introduction to Module
    1. NLP (Natural Language Processing) is an area of artificial intelligence which aids computers understand, interpret and manipulate human language; fill in the gap between human communication and computer understanding.
    2. NLP involves many disciplines, including computer science and computational linguistics;
    3. Two approaches:
      1. Classic by NLTK;
        1. A leading platform for building Python programs to work with human language data;
        2. Available for Windows, Mac OS X, and Linux. NLTK is a free, open source, community-driven project;
      2. AI based: PYTORCH
    4. Lab1 – familiar colab. https://colab.research.google.com/notebooks/welcome.ipynb#scrollTo=gJr_9dXGpJ05
  2. L2 – Categorizing and Tagging Words
    1. Questions
      1. What are lexical categories (parts of speech / word classes)? and how are they used in NLP?
      2. What is a good Python data structure for storing words and their categories?
      3. How can we automatically tag each word of a text with its word class?
    2. Basic NLP techniques: sequence labelling, n-gram models, backoff, and evaluation;
    3. NLP pipeline: Tokenization > Tagging / POS tagging >
    4. The collection of tags used for a particular task is known as a tagset;
    5. Tagger processes a sequence of words and attaches a part of speech tag to each word;
      1. Problem: Many words can be used as either nouns or verbs: Ski, race, address, auction, colour , deal, drain, email, grill, nail, rain…
    6. Tagged Corpora
      1. Several corpora included within NLTK, and they have been tagged for their part-of-speech: (word,tag);
      2. tagged_words() method provided by NLTK to deal with tagged text;
      3. Tagging of words in terms of: Nouns, Verbs, Adjectives and Adverbs;
    7. Python Dictionaries
    8. Automatic Tagging
    9. N-Gram Tagging
    10. Transformation Based Tagging
    11. Determining the Category of Word
    12. Lab2
      1. do exercises 1-10 in chapter 5 of the NLP Processing with Python book (section 5.10)
      2. do some of the other exercises if there is spare time;
  3. L3 – Learning to Classify Text
  4. L4 – Extracting information from textW5 Analysing sentence structure
  5. L5 – bread
  6. L6 – Feed Forward Networks for natural language processing
  7. L7 – ChatBot based on pytorch
  8. L8 – Translation with a Sequence to Sequence Networks and Attention on pytorch
  9. L9 – Generating Names with Character-Level RNN on pytorch
  10. L10 – Classifying Names with Character-Level RNN on pytorch
  11. References
    1. L1 & Lab1
      1. https://www.python.org/downloads/
      2. Jupyter / Pytorch. https://jupyter.org/install
      3. Jupyter Lab. https://jupyterlab.readthedocs.io/en/stable/getting_started/installation.html
      4. Other kernels. https://github.com/jupyter/jupyter/wiki/Jupyter-kernels
      5. TensorFlow. https://www.tensorflow.org/install
      6. Pytorch Errata. http://www.nltk.org/book/
  12. NLP with Python (book study log)
    1. ch1 – Language Processing and Python (20200521)
      1. 1.1 Computing with Language: Texts and Words
      2. 1.2 Closer Look at Python: Texts as Lists of Words
      3. 1.3 Computing with Language: Simple Statistics
      4. 1.4 Back to Python: Making Decisions and Taking Control
      5. 1.5 Automatic Natural Language Understanding
      6. 1.6. Summary
      7. 1.7 Further Reading
      8. 1.8 Exercises
    2. ch2 – Accessing Text Corpora and Lexical Resources (20200521)
      1. 2.1 Accessing Text Corpora
      2. 2.2 Conditional Frequency Distributions
      3. 2.3 More Python: Reusing Code
      4. 2.4 Lexical Resources
      5. 2.5 WordNet
      6. 2.6 Summary
      7. 2.7 Further Reading
      8. 2.8 Exercies
    3. ch3 – Processing Raw Text (20200521)
      1. 3.1 Accessing Text from the Web and from Disk
      2. 3.2 Strings: Text Processing at the Lowest level
      3. 3.3 Text Processing with Unicode
      4. 3.4 Regular Expressions fro Detecting Word Patterns
      5. 3.5 Useful Applications of Regular Expressions
      6. 3.6 Normalizing Text
      7. 3.7 Regular Expressions for Tokenizing Text
      8. 3.8 Segmentation
      9. 3.9 Formatting: From Lists to Strings
      10. 3.10 Summary
      11. 3.11 Further Reading
      12. 3.12 Exercses
    4. ch4 – Writing Structured Programs
      1. 4.1 Back to the Basics
      2. 4.2 Sequence
      3. 4.3 Questions of Style
      4. 4.4 Functions: The Foundation of Structured Programming
      5. 4.5 Doing More with Functions
      6. 4.6 Program Development
      7. 4.7 Algorithm Design
      8. 4.8 A Sample of Python Libraries
      9. 4.9 Summary
      10. 4.10 Further Reading
      11. 4.11 Exercises
    5. ch5 – Categorizing and Tagging Words
      1. 5.1 Using a Tagger
      2. 5.2 Tagged Corpora
      3. 5.3 Mapping Words to Properties Using Python Dictionaries
      4. 5.4 Automatic Tagging
      5. 5.5 N-Gram Tagging
      6. 5.6 Transformation-Based Tagging
      7. 5.7 How to Determine the Category of a Word
      8. 5.8 Summary
      9. 5.9 Further Reading
      10. 5.10 Exercises
    6. ch6 – Learning to Classify Text
      1. 6.1 Supervised Classification
      2. 6.2 Further Examples of Supervised Classification
      3. 6.3 Evaluation
      4. 6.4 Decision Trees
      5. 6.5 Naive Bayes Classifiers
      6. 6.6 Max. Entropy Classifiers
      7. 6.7 Modeling Linguistic Patterns
      8. 6.8 Summary
      9. 6.9 Further Reading
      10. 6.10 Exercises
    7. ch7 – Extracting Information from Text
    8. ch8 – Analyzing Sentence Structure
    9. ch9 – Building Feature-Based Grammars
    10. ch10 – Analyzing the Meaning of Sentences
    11. ch11 – Managing Linguistic Data

 

NLP

電工技術20190918

Gallon WongLeave a comment
  1. 簡介
    1.  初階(工聯會20190918-12課)
      1. L1-L6 理論筆記
      2. L7-L12 實習操作(自備工具及已代購材料)
    2. 中級
  2. 初階(By K.L. Poon)
    1. L1
      1. 電力的產生
        1. 原子/原子核/電子/質子/中子
        2. 電子流(負到正)/電流(正到負)
        3. 摩擦產生的電動力(物質在壓力下, 電子轉至另一物質表面, 形成帶正電荷/負電荷的物質)
        4. 熱力產生的電動力
        5. 化學作用產生的電動力
        6. 光生電力
        7. 壓力產生的電動力
        8. 磁感產生的電動(導線和磁力線相割)
      2. 直流電(DC)和交流電(AC)
        1. 交流電(AC): 具有週期性的變化;
        2. 高壓電: AC>1000V, DC>1500V;
        3. 直流電(DC)產生
          1. 乾電池(Dry cell): 一次電池(碱性 1.6-1.7v / 碳性 1.5v)
          2. 可蓄電(Re-chargeable cell): 鉛酸蓄電池 / 鋰電池;
      3. 其他
        1. 大膽假設, 小心求証, 不恥下問;
        2. 電力祖師 法拉第;
        3. 電子祖師 富蘭克林;
    2. L2
      1. 直流電(DC)和交流電(AC)
        1. 直流電(DC)產生
          1. 電源供應器(Power supply) / 交流轉直流變壓器(火牛): 具備初級/次級功能;
          2. 直流發電機(D.C generator)
        2. 交流電(AC)產生: 直流電升壓后在傳輸的損耗大, 不及交流電的按現
          1. 交流發電機(A.C generator)
      2. 不同種類的發電方法
      3. 電學基本認識
        1. V=IR
        2. P=VI
    3. L3
      1. 交流電的電壓與電流值
      2. 電路的危險性
    4. L4
      1. 電動機
      2. 變壓器
    5. L5
      1. 光管結構
  3. 參考
    1. 法例及標準
電工技術20190918