Previous: Limit (8) technology: integration and separation
Contents :
** 0x01 color space
** 0x02 best books on machine learning
how to read a scientific research ** 0x03 News
** Software 0x04 Why the Take the Projects Within last you longer of Think - A Statistical Model
** 0x05 database basis
** 0x06 State- The-Art-of
** Manager 0x07
** 0x08 Founder apos Developer Tools A Guide to the Business Building
0x01 color space
原文: color-spaces
This article through interaction and commentary section in the page, only to explain the concept of mutual conversion between the concept of color space, and the different color space in the RGB color model. Graphic interaction is superb, recommended reading text.
Brief annotation :
A RGB color space, is constituted by RGB. Intensity can be converted to RGB by the equation, for example:
- Linear intensity: I = (intensity value) = (encoded value of R, G, B)
- Non-linear intensity: I = (intensity value) = (encoded value of R, G, B) ^ 2.0
The intensity of a color space conversion formula, the same color, for example, from light to dark gray, human eye to see the difference are different. This difference is called the tone response curve (TCR).
Why should I choose a different intensity conversion formula for it?
Because different intensity conversion formula different TCR formed.
Thus it will form a different color space.
Why not just use a linear transformation of it?
Because the human eye's perception of light is non-linear.
Different colors choose different comfort of TCR can be obtained perception.
The different conversion equation determines a different color space.
It can be transformed into another color space between them.
Suppose color space A, and the color space B. A color space can be calculated then there are R, G, B color space by R indicates how the B, G, B:
R(A) = w11* R(B) + w12 * G(B) + w13* B(B)
G(A) = w21* R(B) + w22 * G(B) + w23* B(B)
B(A) = w31* R(B) + w32 * G(B) + w33* B(B)Thus, according to Glassman principles:
Two color space in the same shade of color CA, CB. Each of the other two colors are mixed in the same shade of color space CA ', CB'. The resulting shade or in two different colors.
Thus, W can be obtained from the color space conversion matrix into a color space B of A:
w11 w12 w13
w21 w22 w23
w31 w32 w33In turn, required color space transformation matrix from A to B in the color space, the inverse matrix S can only requires:
s11 s12 s13
s21 s22 s23
s31 s32 s33That is to say WxS=I.
Color space of R, G, B values are [0-255], the formation of a color space cube.
A conversion from the color space into a color space B, then A may be a point in space to which B is not in the color space cut into cubes.
Common RGB color space:
- sRGB color space
- XYZ color space
A common model for the luminance (brightness) and chrominance (color) are separated, it is called the YCbCr, the conversion formula is as follows:
Y = 0.299R + 0.587G + 0.114B
Cb = 0.564(B - Y)
Cr = 0.713(R - Y)Conversely, conversion from YCbCr to RGB:
R = Y + 1.402Cr
B = Y + 1.772Cb
G = Y - 0.344Cb - 0.714Cr[. 1] Pastel IS A to Generate Command-Line Tool, the Analyze, Convert Colors Manipulate and
[2] a gradual Video Technologies
0x02 best machine learning books
原文： best-deep-learning-books-updated-for-2019
If you plan to study machine learning system, two paths, one is starting from the books. Another is to start from the tool chain. Books start and there are books from more than vast, how to choose the questions, this blog gives TopN recommendation.
- Deep Learning by Ian Goodfellow, Yoshua Bengio, Aaron Courville
- Grokking Deep Learning by Andrew W. Trask
- Deep Learning with Python by Francois Chollet
- Hands-On Machine Learning with Scikit-Learn and TensorFlow by Aurélien Géron
- The Hundred-Page Machine Learning Book by Andriy Burkov
- Reinforcement Learning: An Introduction (2nd Edition) by Richard S. Sutton, Andrew G. Barto
- Deep Reinforcement Learning Hands-On by Maxim Lapan
- Learning From Data by Yaser S. Abu-Mostafa, Malik Magdon-Ismail, Hsuan-Tien Lin.
- The Book of Why by Judea Pearl, Dana Mackenzie.
- Machine Learning Yearning by Andrew Ng.
- Interpretable Machine Learning by Christoph Molnar.
- Neural Networks and Deep Learning by Michael Nielsen.
0x03 How to Read a science news
- Where did the story come from?
- What kind of research was this?
- What did the research involve?
- What were the basic results?
- How did the researchers interpret the results?
- Conclusion
[1] https://www.nhs.uk/news/cancer/drinking-very-hot-tea-linked-risk-1-type-oesophageal-cancer/
0x04 Why software projects take longer than you think – a statistical model
From a statistical point of view why the estimates are always longer than the actual time, there is some truth. In fact, it is very difficult to estimate time inside, we can not even list the unit simple and time-consuming to do accumulate.
0x05 database infrastructure
People always forget the great role database software in the world to bring computer technology inside, database technology plays a very important one seat, database technology is behind the relational algebra, although we later learned SQL, SQL itself is more intuitive than relational algebra.
0x06 State-of-the-Art
Original : https://en.wikipedia.org/wiki/State_of_the_art
The term State-of-the-Art represents the highest level of the current of a field study.
State-of-the-art (sometimes cutting edge or leading edge) refers to the highest level of general development, as of a device, technique, or scientific field achieved at a particular time.
0x07 Manager
Say "manager" on a computer meaning, manager generally occurs in the presence of N objects of the same kind of situation, if only the object code of a certain type, you do not need it to do manager, for example mysqlClient, this client It will be only one, nothing good "manager" of. But if you have some session, the session will need to be manager, you will naturally have a SessionManager. But the manager also has strong and weak points, if only to hold session, simply add / remove, to do batch is weak manager. And if you want to do a bunch of session scheduling strategy, it is a little heavy point, for example, to do garbage collection, empty overtime, do LRU removed and the like, to moderate a little bit. Manager have different data structures, different data structures will bring a different insert, search efficiency, without structural Manager is designed without thinking, inefficient, Manager of the unit is always expanding, and even only a mere 3-4 units without proper design, there are also unable to accommodate growth complexity. In the computer, we need to spend a lot of intelligence in control complexity.
0x08 Founder's Guide to Building a Developer Tools Business
- Understanding the developer persona:
- The most valuable features are the ones that automate the dull stuff and do what they say they do
- Offer multiple/tiered pricing models
- They look at the pricing model of your product to see if you offer a free tier so that they can test your platform without investing anything
- Offer easy-to-navigate docs
- Docs should be detailed, easy to read, and packed with examples
- Offer demos, previews, and trials
- You can still offer demos, trials, or quick screencasts to showcase your product.
- Offer integrations and CLI tools
- CLI tools and language libraries enhance your chances of success in developers’ circles
- Offer community support
- Word spreads fast if users can get support and resolve their issues quickly, and your product becomes more and more popular
The reason that it know everything, but we always hard to be strictly enforced, and the six easy steps to achieve the ultimate, it is difficult to see the complete implementation of the policy, so that we can always do things the ordinary the level of excellence that is difficult to output software. The general level of policy implementation efficiency, so we can not get 10x growth in the software manufacturing, so that we ourselves will be disappointed. There are many people is difficult to grasp the force, one of which is execution.
--end--