關於 classification example ，我們在網路上蒐集到這些相關的討論、資訊與評價

🤓 น้องๆ รู้ไหมครับว่า computer vision
อ่านว่า "คอมพิวเตอร์วิชั่น"
เป็นศาสตร์ทางคอมชั้นสูงอีกแขนงหนึ่ง
ที่ทำให้คอมพิวเตอร์มองเห็นได้ดุจดังตามนุษย์
เป็นไงฟังแล้วเท่ห์มั๊ยละ
.
ปกติมนุษย์เวลาเห็นวัตถุ สิ่งของ สิ่งมีชีวิต อยู่ในรูป หรือในวีดีโอ
เช่น เห็นปูุ๊บก็จะบอกได้ทันทีว่า
ในรูปหรือวีดีโอมีหมา แมว แก้วน้ำ แจกัน ฯลฯ
เรื่องแบบนี้สำหรับคน ถือว่าโคตรรรรรรรรรง่ายๆ
.
แต่ทว่าในมุมของคอมพิวเตอร์
มันยากมากกกกกกกกกกกกกก
.
ถ้าคอมมันทำได้ก็ถือว่าฉลาดมากแล้ว
ถือว่า intelligent เป็นสิ่งอัจฉริยะในมุมของคอม
ซึ่งการแยกแยะว่าสิ่งที่อยู่ในรูปว่ามันเป็นอะไร
ภาษาอังกฤษเรียกว่า classification (แยกแยะประเภท)
.
สำหรับศาสตร์ computer vision 💻 💻
มันแตกแขนงได้หลายเทคนิค หลายสาย
แต่ในโพสต์นี้จะพูดถึงเฉพาะแค่ 2 เรื่อง เท่านั้นได้แก่
.
☑ 1) การทำ Object detection
เป็นการตรวจจจับวัตถุในรูป
ในมุมการ classification มันแค่บอกว่าสิ่งในรูปคืออะไร
แต่ Object detection จะไม่ใช่แค่บอกว่าสิ่งที่อยู่ในรูปคืออะไรเท่านั้น
.
😲 😲 มันเหมือนอัพเกรตจากความสามารถ classification
เพื่อพัฒนาไปตีกล่องสี่เหลี่ยมล้อมรอบ (ภาษาอังกฤษเรียกว่า Box)
ตีกล่องเพื่อบอกตำแหน่งสิ่งที่มันตรวจจับในรูปหรือในวีดีโอได้ ...โอ้วแม่เจ้าจะเทพไปถึงไหน
.
.
☑ 2) การทำ Segmentation
จะแอดวานซ์จาก Object detection
เหมือนเป็นเวอร์ชั่นอัพเกรตอีกต่อหนึ่ง
.
คือมันไม่ได้ตีกรอบล้อมรอบวัตถุนะ
แต่มันจะยิ่งกว่านั้น
...ก็คือมันจะไฮไลท์ (ฝรั่งบอกว่า mask)
....มันจะระบุพื้นที่วัตถุ สิ่งของ ที่มันตรวจจับได้ (เหมือนเราแรงงานไปยังวัตถุนั้น ๆ)
.
.
ซึ่งศาสตร์ computer vision ปัจจุบันนี้
นิยมเอา AI (ปัญญาประดิษฐ์) มาใช้งาน
จริงมันก็ทำอย่างอื่นได้นอกจากสองเรื่องที่ผ่าน เช่น
pose estimation ตรวจจับจุดต่างๆ ของมนุษย์ แบบกล่องสองผี 💀💀 💀 ....จนปรึ้นๆๆ น่ากลัวหรือเปล่า เป็นต้น
.
และมีข่าวดีสำหรับใครที่สนใจเขียนโปรแกรม 👏 👏
สำหรับงาน Object Detection กับ Segmentaion
ผมทำตัวอย่างการใช้งานไว้แล้ว
มี 5 แบบ ลองเข้าไปดูโค้ดได้ที่นี้
.
🔥 1. การช้งานไลบรารี่ imageai ตรวจจับวัตถุในรูป
https://colab.research.google.com/…/1uQnZfPlRhplvcZKWiXn1je…

🔥 2. การใช้งานไลบรารี่ pixellib ตรวจจับวัตถุในรูป
https://colab.research.google.com/…/1llWzReE3rS9wDfSGGm8M7R…

🔥 3. การใช้ไลบรารี่ Tensorflow ตัวอย่างของ Google ตรวจจับวัตถุในรูป
https://colab.research.google.com/…/12K-4uQ1tAvOukLb1-lwXx4…

🔥 4. การใช้ไลบรารี่ Detectron2 ของเฟสบุ๊ค ตรวจจับวัตถุในรูป
https://colab.research.google.com/…/1jnWFADFdZHz1LSyfXVKHY3…

🔥 5 ตัวอย่างการใช้โมเดล Msk RCNN ตรวจจับวัตถุในรูป
https://colab.research.google.com/…/1JGRIMQ1YSdMXkEZdC6QNGb…

.
ปล. มีแต่โค้ด python เพราะต้องทำใจอะนะ
เพราะงานด้านนี้ ภาษานี้คืออันดับหนึ่งในงาน AI ณ ห่วงเวลานี้
(สามารถไปอ่านตำราเรียน ม.1-2-3 มีเนื้อหา python อยู่ในตำรา)
.
.
+++++ขอประชาสัมพันธ์ (ขายของ)
📔 หนังสือ "ปัญญาประดิษฐ์ (AI) ไม่ยาก" เข้าใจได้ด้วยเลขม. ปลาย (เนื้อหาภาษาไทย)
.
ถ้าสนใจสั่งซื้อเล่ม 1 ก็สั่งซื้อได้ที่ (เล่มอื่นๆ กำลังทยอยตามมา)
👉 https://www.mebmarket.com/web/index.php…
.
ขออภัยยังไม่มีเล่มกระดาษจำหน่าย มีแต่ ebook
.
ส่วนตัวอย่างหนังสือ ก็ดูได้ลิงค์นี้
👉 https://www.dropbox.com/s/fg8l38hc0k9b…/chapter_example.pdf…
.

✍เขียนโดย โปรแกรมเมอร์ไทย thai programmer
🤓 Do you know that computer vision?
Read ′′ Computer Vision ′′
It's another elite computer science.
That makes computers visible as human eyes.
How are you? Are you cool?
.
Normally humans see objects of living things in the photo or video.
Such as seeing a crab, I can tell you right away.
In the picture or video, there are dogs, cats, mug, vase etc.
This kind of thing for people is considered simple.
.
But in the corner of computer
It's very difficult.
.
If a computer can do it, it's very clever.
Intelligent is considered genius in the corner of the computer.
Which digest what's in the picture is what it is.
English is called classification (type digest)
.
For computer vision science 💻 💻
It's broken. Many tricks. Many lines.
But in this post, it's only about 2 stories.
.
☑ 1) Object detection
Detection of the objects in the photo
In classification angle, it just says what is in the picture.
But Object detection won't just say what's in the picture.
.
😲 😲 It's like an upgrade from ability classification.
To develop, hit a square box, surround (English is called Box).
Hit the box to spot what it detects in the photo or video... Oh my god how can it be?
.
.
☑ 2) Making Segmentation
Will admin from Object detection
It's like another upgrade version.
.
Well, it doesn't hit the frame surrounding the object.
But it will be more than that.
... Well, it's going to be highlighted (Foreigner says mask)
.... It will identify the areas, objects, things that they detect (like we labor to that object).
.
.
The science of computer vision nowadays.
Popular to use AI (Artificial Intelligence)
True, it can do anything else but two things that pass, e.g.
pose estimation detects different points of human in two ghost box. 💀💀 💀 💀 💀.... so bad, scary, etc.
.
And some good news for anyone who is interested in programming 👏 👏
For Object Detection with Segmentaion
I made an example of use.
There are 5 designs. Check out the code here.
.
🔥 1. Imageai library work. Detection of objects in the photo.
https://colab.research.google.com/drive/1uQnZfPlRhplvcZKWiXn1jeytJIFEVLkV

🔥 2. usage of pixellib libraries. Detect objects in the photo.
https://colab.research.google.com/drive/1llWzReE3rS9wDfSGGm8M7RQ25jeEfSIi

🔥 3. uses of Tensorflow's Tensorflow library. A sample of Google detects objects in the photo.
https://colab.research.google.com/drive/12K-4uQ1tAvOukLb1-lwXx4bnXkeQupTk

🔥 4. Facebook's Detectron2 Library Uses rūp Objects Detector
https://colab.research.google.com/drive/1jnWFADFdZHz1LSyfXVKHY3fIwuY5F_uo

🔥 5 examples of using Msk RCN model. Detect objects in the photo.
https://colab.research.google.com/drive/1JGRIMQ1YSdMXkEZdC6QNGbI722tEQJTE

.
Ps. There are only code python because I need to get over it.
Because of this side job, this language is number one in AI event. At this time, I'm
(Can go to read textbooks. 1-2-3 python content in the textbook)
.
.
+++++ Public relations (selling items)
📔 The book ′′ Artificial Intelligence (AI) is not difficult It can be understood by the number. Tipping (Thai language content)
.
If you are interested in ordering book 1, you can order at (other books are gradually following).
👉 https://www.mebmarket.com/web/index.php?action=BookDetails&data=YToyOntzOjc6InVzZXJfaWQiO3M6NzoiMTcyNTQ4MyI7czo3OiJib29rX2lkIjtzOjY6IjEwODI0NiI7fQ&fbclid=IwAR11zxJea0OnJy5tbfIlSxo4UQmsemh_8TuBF0ddjJQzzliMFFoFz1AtTo4
.
Sorry, no paper books available. Only ebooks.
.
Personal like the book. You can see this link.
👉 https://www.dropbox.com/s/fg8l38hc0k9b0md/chapter_example.pdf?dl=0
.

✍ Written by Thai programmer thai coderTranslated

[ NO VICTORY FOR HONG KONG (yet) ] To my friends outside of Hong Kong,

You may have heard that the protestors have won: the government will withdraw the extradition bill, and we've gotten what we want. THIS IS NOT TRUE. We have five demands for the administration since the beginning, and retracting the bill at this time is too little, too late. The protests will continue, and this is why:

We protested because the bill would make everybody a potential victim of Mainland China's deeply unjust legal system. But what we've seen from the government's actions since June, is that the rule of man has already taken over in Hong Kong. The things we wanted to prevent in our city is already happening, bill or no bill:

FOR EXAMPLE: You can be tackled to the ground by the police just for shouting "Where is your conscience?" across the street. If you're arrested, you will be tortured, and forcibly undressed and humiliated by a cop while many more stand by and watch. You can be fired for supporting the protest on Facebook due to pressure from Beijing. If you're a student, you may have the cops called on you and your name submitted to the Education Bureau, just for wearing a mask as a sign of protest.

None of those things have made international headlines, but they are just a few of the countless acts of terror that happened within a span of ~90 days.

That is why withdrawing the bill at this stage changes nothing. Without a systemic change, and measures to hold the police and administration accountable, Hong Kong's problems will persist. And so the protests will go on. We do it because we can't let Hong Kong sink any lower.

Fight for Freedom, Stand with Hong Kong.

The Five Demands:
1. Full withdrawal of the Extradition Bill
2. An independent commission of inquiry into alleged police brutality
3. Retracting the classification of protesters as “rioters”
4. Amnesty for arrested protesters
5. Dual universal suffrage (for both the Legislative Council and the Chief Executive)

ในวิชา "วิทยาการคำนวณ" ระดับชั้น ม. 5
ได้ดึงวิชา data science (วิทยาศาสตร์ข้อมูล)
มาปูพื้นฐานให้เด็กๆ ได้เรียนกันแล้ว นับว่าเป็นโชคดี
เพราะวิชาพวกนี้เป็นของสูง กว่าจะสัมผัสก็คงตอนป.ตรี โท เอก
...Continue ReadingIn the subject of ′′ Calculation Theology ′′ class. 5
Pulled data science (data science)
Let's master the foundation for kids to learn. It's considered lucky.
Because these subjects are high to touch. It's probably in the middle of the year. Tri To Aek

Which I will review the content to read roughly. The content is divided into 4 chapters.
.
👉 ++++ Chapter 1-Information is valuable +++++
.
Data science in the textbook. Used by Thai name as ′′ Information Science ′′
This chapter will mention Big Data or big data with lots of valuable information.
And so much role in this 4.0 s both public and private sector.
.
If you can't imagine when you played Google search network, you'll find a lot of information that you can use in our business. This is why data science plays a very important role.
.
It's not surprising that it makes the Data Scientist s' career (British name data scientist) play the most important role and charming and interesting profession of the 21th century.
.
Data science, if in the book, he defines it
′′ Study of the process, method or technique to process enormous amounts of data to process to obtain knowledge, understand phenomena, or interpret prediction or prediction, find out patterns or trends from information.
and can be analysed to advise the right choice or take decision for maximum benefit
.
For Data science work, he will have the following steps.
- Questioning my own interest.
- Collect information.
- Data Survey
- Data Analysis (analyze the data)
- Communication and Results Visualization (Communicate and visualize the results)
.
🤔 Also he talks about design thinking... but what is it?
Must say the job of a data scientist
It doesn't end just taking the data we analyzed.
Let's show people how to understand.
.
The application design process is still required.
To use data from our analytics
The word design thinking is the idea. The more good designer it is.
Which Data Scientists Should Have To Design Final Applications
Will meet user demand
.
👉 ++++ Chapter 2 Collection and Exploration +++++
.
This chapter is just going to base.
2.1 Collection of data
In this chapter, I will talk about information that is a virtual thing.
We need to use this internet.

2.2 Data preparation (data preparation)
Content will be available.
- Data Cleaning (data cleansing)
- Data Transformation (data transformation)
In the university. 5 is not much but if in college level, you will find advanced technique like PCA.
- Info Link (combining data)

2.3 Data Exploration (data exploration)
Speaking of using graphs, let's explore the information e
Histogram graph. Box plot diagram (box plot). Distributed diagram (scatter plot)
With an example of programming, pulls out the plot to graph from csv (or xls) file.

2.4 Personal Information
For this topic, if a data scientist is implementing personal data, it must be kept secret.
.
Where the issues of personal information are now available. Personal Data Protection is Done
.
.
👉 ++++ Chapter 3 Data Analysis ++++
.
Divided into 2 parts:
.
3.1 descriptive analysis (descriptive analytics)
Analyzing using the numbers we've studied since
- Proportion or percentage
- Medium measurement of data, average, popular base.
Correlation (Correlation) relationship with programming is easy.
.
.
3.2 predictive analysis (predictive analytics)
.
- numeric prediction is discussed. (numeric prediction)
- Speaking of technique linear regression, a straight line equation that will predict future information.
Including sum of squared errors
Let's see if the straight line graph is fit with the information. (with programming samples)
- Finally mentioned K-NN (K-Nearest Neighbors: K-NN) is the closest way to finding K-N-Neighborhood for classification (Category)

*** Note *****
linear regression กับ K-NN
This is also an algorithm. One of the machine learning (machine learning, one branch of AI)
Kids in the middle of the day, I get to study.
.
.
👉 +++ Chapter 4 Making information pictured and communicating with information +++
.
This chapter doesn't matter much. Think about the scientist after analyzing what data is done. The end is showing it to other people by doing data visualization. (Better summoning)
.
In contents, it's for example using a stick chart, line chart, circular chart, distribution plan.
.
The last thing I can't do is tell a story from information (data story telling) with a message. Be careful when you present information.
.
.
.
*** this note ***
😗 Program language which textbooks mentioned and for example.
It's also python and R language
.
For R language, many people may not be familiar.
The IT graduate may be more familiar with Python.
But anyone from the record line will surely be familiar.
Because R language is very popular in statistical line
And it can be used in data science. Easy and popular. Python
.
But if people from data science move to another line of AI
It's deep learning (deep learning)
Python will be popular with eating.
.
.
#########
😓 Ending. Even I wrote a review myself, I still feel that.
- The university. 5 is it going to be hard? Can a child imagine? What did she do?
- Or was it right that I packed this course into Big Data era?
You can comment.
.
But for sure, both parents and teachers are tired.
Because it's a new content. It's real.
Keep fighting. Thai kids 4.0
.
Note in the review section of the university's textbook. 4 There will be 3 chapters. Read at.
https://www.facebook.com/programmerthai/photos/a.1406027003020480/2403432436613260/?type=3&theater
.
++++++++++++++++++++
Before leaving, let's ask for publicity.
++++++++++++++++++++
Recommend the book ′′ Artificial Intelligence (AI) is not difficult ′′
It can be understood by the number. End of book 1 (Thai language content)
Best seller ranked
In the MEB computer book category.
.
The contents will describe Artificial Intelligence (A) in view of the number. The end.
Without a code of dizzy
With colorful illustrations to see, easy to read.
.
If you are interested, you can order.
👉 https://www.mebmarket.com/web/index.php?action=BookDetails&data=YToyOntzOjc6InVzZXJfaWQiO3M6NzoiMTcyNTQ4MyI7czo3OiJib29rX2lkIjtzOjY6IjEwODI0NiI7fQ&fbclid=IwAR11zxJea0OnJy5tbfIlSxo4UQmsemh_8TuBF0ddjJQzzliMFFoFz1AtTo4
.
Personal like the book. You can see this link.
👉 https://www.dropbox.com/s/fg8l38hc0k9b0md/chapter_example.pdf?dl=0
.
Sorry, paper book. I don't have it yet. Sorry.
.
✍ Written by Thai programmer thai progammerTranslated

รีวิว Casio Pro Trek F20A นาฬิกาที่แข็งแรงที่สุดจาก Casio ผ่านการทดสอบระดับเกรดทหาร

Casio Pro Trek F20A เปิดราคาเมืองไทย 14900 บาท เท่านั้น สามารถไปชมตัวจริงได้ที่งาน International Watch Fair วันที่ 21 สิงหาคม 2561 - 30 กันยายน 2561 นี้ ที่เซนทรัลชิดลม กรุงเทพมหานคร

Wear OS by Google™
4 GB internal memory
512 MB RAM
512 MB of RAM available.
Wi-Fi (IEEE 802.11 b/g/n)
Bluetooth® Smart
Super-Illuminator
MIL-STD-810G compliant
The MIL-STD-810 is a U.S. military technical standard that specifies environmental test conditions for military equipment. The standard defines tests for the compatibility of equipment and devices with extreme temperatures and air pressures, solar radiation, humidity, chemicals, acceleration forces and vibrations.
Smartphone time
Double LCD (320 x 300)
Touchscreen
Sunrise/sunset display
Fishing Timer
A special sensor measures the air pressure (measuring range: 260/1100 hPa) and presents this on the display in the form of a symbol. This enables early detection of weather trends.
Digital-compass
A built-in direction sensor detecs the magnetic north.
10,000 m Altimeter
A pressure sensor detects changes in the air pressure and converts the result into an altitude up to 10,000 m.
Altimeter data memory
Depending on the model, up to 40 sets of elevation data can be stored in the watch's data memory and re-accessed at any time. Each of these data sets consists of a measured altitude as well as the date and time. In addition to the individual altitudes, the maximum and minimum altitudes during a measurement are also stored.
Moon Phase Indicator
The moon phase indicator shows the current moon phase based to the earth
Tide indicator
Once the high tide time has been entered for a selected location, the watch can display the low and high tide phases at this location on any given date. This can be quite useful, for example, if you want to know the patterns of the tides at your destination in advance.
Gyrometer
Ideal for outdoor activities: The gyro sensor shows you every little change in direction during navigation, ensuring you always reach your destination. The sensor detects the smallest movements of the watch and acts against this by supporting optical image stabilisation or the alignment of the display.
Acceleration sensor
The motion sensor responds to acceleration and shows this as a speed on the display. The sensor also supports the auto rotation function of the watch, ensuring that the image is always displayed the right way up on the display.
GPS function
Measures your current position using artificial satellites circling the globe.
World time function
Displays the current time in major cities and specific areas around the world.
Daily alarm
The daily alarm reminds you of events that recur every day, by emitting an audible signal at the set time.
Snooze feature
Each time you stop the alarm, it sounds again after a few minutes.
Vibration alert
The vibration alarm discreetly reminds you of important appointments, without emitting a loud signal.
Airplane Mode (Offline Mode)
At the touch of a button, the watch switches to offline mode. This means that you can disable the wireless or GPS receiver when in the air for the duration of the flight.
Microphone
Make your daily life easier using practical voice commands spoken directly into the microphone on your watch. For example, you can start a Google search by saying "OK Google" — without the need to touch the display, or even take your smartphone out of your pocket. Many actions can be performed via the microphone using voice commands.
Mineral glass
The tough mineral glass resists scratching.
Resin case
Resin band
Synthetic resin is the ideal material for wrist straps thanks to its extremely durable and flexible properties.
Battery level indicator
An indicator shows the current battery level.
Water resistance classification (5 bar)
This watch can be worn when showering and bathing - it is tested for water resistance up to 5 bar according to ISO 22810.
Dimensions (H x W x D)
61,7mm x 57,7mm x 15,3mm
Weight
approx. 90 g
Availability of models and/or lineups may differ depending on the country or region.
Technical amendments and errors excepted. The sizes in figures do not correspond to the original sizes.
The colours may differ slightly from the original.
ติดตามพวกเราต่อได้ที่
www.facebook.com/stepgeektv
lifeisgadget@gmail.com
https://shopee.co.th/stepgeekstore
www.stepgeek.tv
LINE @ : @lyb1717r ต้องมี @ ด้วยนะ
โทร 095-9642699 ติดต่อซื้อมือถือ หรือ ติดต่อสินค้ารีวิว
เบอร์หน้าร้าน 098-4466914 มาบุญครอง ชั้น 4

00:00 In today's episode, I have a question from Umang, who asked, "Hey Rayner, how do I scan the stock markets for my trading setup? Because there are so many stocks out there. Hundreds, if not thousands of stocks."

00:17 So my response to this is that there's a four-step procedure that you can follow, number one: finding a trading setup or a pattern is just like finding a wife.

00:37 if you are looking for a wife who is a brunette, Asian, Chinese, Indian, or whatsoever, you have to know what you're looking for in the first place, and it's the same for trading the markets.

00:54 Are you looking for a trending stock? a breakout stock, or maybe a stock that's tanked over the last twelve months and you're going to buy near the bottom whatsoever.

01:10 Once you've defined your trading setup then you can go into step two: Determine the market cap of the stocks you want to trade. For example, for those of you who are in the U.S., you have Russell 3000. Russell 3000 comprises of the large-cap, mid-cap, and the small-cap stocks. So, if you can define what market capitalization of stocks you want to trade, you can filter it down further. But if you're let's say, in Singapore, there's the large-cap, mid-cap, you have a few thereabout. You don't really need to do this classification, it's only for those of you who are in countries where you have thousands and thousands of stocks.

01:48 So once you have classified the market cap at step two, proceed to step three: Scan the markets. Let's say you want to scan stocks that are trending, what you want to do is to put up the twenty, fifty and two hundred period moving average, so if the stock is above the three moving averages, chances are it's a trending stock. You can also go through your entire watchlist and scan all the stocks which are above these moving averages. Alternatively, if you want to trade breakout stocks, you can just look for stocks that are on a 52-week high, breaking out of their 200-day high, etc. Another example I can give you is that if you want to buy stocks that are doing very bad over the last 12 months and you want to catch bottoms, what you can do is to scan stocks at their 52-week low, 200-day low, and etc.

02:44 Once you've scanned the markets, what you want to do is do number four: Put it all in a spreadsheet. Let's say you have 500 stocks to scan, you can do it all on the weekend. You want to put down those stocks in a spreadsheet that are close to having their setup occur so that when the market opens next week, you will only focus on those stocks in your spreadsheet that are about to have their setup occur. So, what you want to do is to find the gems among your 500 stocks and put it into a spreadsheet, and focus on it this coming week.

03:40 Now that you can scan the stock market and find the trading setups that you want, some of you might be wondering if there is any software to do this work for you? The only one that I do know of is Finviz, F-I-N-V-I-Z, and that is only for the U.S. stock market. They have this kind of quick scanner on whether you want to use fundamentals or technicals. As for local Singaporean stocks, Philippine stocks, Indian stocks, I do not know of such tools that are in the market right now. But for the U.S. there is Finviz. If there are no scanners you can just go through the four steps I shared with you. Number one, define your trading setup. Number two, know the market cap that you want to trade. Number three, scan the markets. And number four, put those stocks that you have scanned into a spreadsheet so you can track them when the market opens this coming week.

04:35 So that's all I have for you in this week's episode of Ask Rayner. If you're looking for proven trading strategies that work, you can go down to my website tradingwithrayner.com and access my materials for free. If you have any questions, let me know in the comments section below, and if you have enjoyed this video, hit the Like button and I would really appreciate it. I'll talk to you soon.

If you want more actionable trading tips and strategies, go to https://www.tradingwithrayner.com

Thanks for watching!

FOLLOW ME AT:
Facebook: https://www.facebook.com/groups/forextradingwithrayner
Twitter: http://www.twitter.com/rayner_teo
My YouTube channel: http://bit.ly/2EFg5VN