搜尋此網誌

2026年5月28日星期四

Dictionary Comprehensions in Python

Revision:

In Python, lists, tuples, sets, and dictionaries are all built-in data structures, but they differ in mutability, ordering, and how they store data. Lists and dictionaries are mutable, tuples are immutable, and sets enforce uniqueness. Dictionaries store key–value pairs, while the others store single values.

List: [ ]

Tuple: ( )

Set: { } or set()

Dictionary: {key: value}


myList = [1,2,3,4,5]

[2*item for item in myList]

This a list comprehension, not a for loop.

In a for ... in loop, the colon (:) is mandatory because it introduces a block of indented code.

for item in myList:

    print(2*item)


We can use any valid variable name instead of "item" in a for ... in loop or a list comprehension. The name is just a placeholder that represents each element in the sequence while iterating.

What is a Key–Value Pair?

A dictionary is like a real-world dictionary:
Key = the word you look up.
Value = the definition you get.

In Python:
Key must be unique and immutable (string, number, tuple).
Value can be anything (string, number, list, another dict, etc.).

person = {
    "name": "Alice",
    "age": 25,
    "city": "Hong Kong"
}

"name" → key, "Alice" → value
"age" → key, 25 → value
"city" → key, "Hong Kong" → value

animalList = [('a', 'anteater'), ('b', 'bear'), ('c', 'cat'), ('d', 'dog')]
animals = {item[0]: item[1] for item in animalList}
animals

List of tuples
[('a', 'anteater'), ('b', 'bear'), ('c', 'cat'), ('d', 'dog')]

Each tuple has two elements:
item[0] → the key (like 'a')
item[1] → the value (like 'anteater')

Dictionary comprehension
{item[0]: item[1] for item in animalList}

Loops through each tuple in animalList.
Uses the first element (item[0]) as the key.
Uses the second element (item[1]) as the value.

{'a': 'anteater', 'b': 'bear', 'c': 'cat', 'd': 'dog'}

>>> animalList = [('a', 'anteater'), ('b', 'bear'), ('c', 'cat'), ('d', 'dog')]
>>> animals = {key: value for key, value in animalList}
... animals
...
{'a': 'anteater', 'b': 'bear', 'c': 'cat', 'd': 'dog'}

animals.items()
dict_items([('a', 'anteater'), ('b', 'bear'), ('c', 'cat'), ('d', 'dog')])
Asking Python to give you all the key–value pairs in the dictionary.

list(animals.items())
[('a', 'anteater'), ('b', 'bear'), ('c', 'cat'), ('d', 'dog')]

A list comprehension that builds a list of dictionaries from your animals dictionary
[{'letter': key, 'name': value} for key, value in animals.items()]

animals.items()
Produces key–value pairs like:
('a', 'anteater'), ('b', 'bear'), ('c', 'cat'), ('d', 'dog')

Looping with for key, value in animals.items()
Each iteration unpacks the tuple into key and value.

Dictionary inside the comprehension
For each pair, you create a new dictionary:
{'letter': key, 'name': value}

Result:
[{'letter': 'a', 'name': 'anteater'}, {'letter': 'b', 'name': 'bear'}, {'letter': 'c', 'name': 'cat'}, {'letter': 'd', 'name': 'dog'}]

Microsoft Copilot

Startup Company

bear down on: to move quickly toward someone or something in a determined way

get something off the ground: to start happening successfully; to make something start happening successfully

payroll: a list of people employed by a company showing the amount of money to be paid to each of them

mundane: not interesting or exciting

monte: a card game

viable: that can be done; that will be successful

topology: a branch of mathematics concerned with those properties of geometric configurations (such as point sets) which are unaltered by elastic deformations (such as a stretching or a twisting) that are homeomorphisms

disabuse somebody (of something) to tell somebody that what they think is true is, in fact, not true

Bill Gates "Source Code"

Online Dictionaries Used:

hk.dictionary.search.yahoo.com

www.oxfordlearnersdictionaries.com

www.merriam-webster.com

2026年5月26日星期二

西九文化區日落美景

 

Aperture Priority (光圈先決): f/9

ISO: 100

(Shutter Speed) 快門: 1/100 second

Focal Length (焦距): 135mm

Equivalent Focal Length (等效焦距): 216mm

Tripod Used and Image Stabilizer Off

White Balance (白平衡): Shade (陰影模式)

效果:相機會自動加暖色調,橘色比肉眼看更明顯

負曝光補償(Exposure Compensation): -1ev

Landscape Picture Style (Contrast: +1, Saturation: +1)

Evaluative Metering (權衡式測光)

Safety Precautions of Tripod Usage

Manfrotto MKBFRTA4GT-BH

Designed for the advanced travel photographer, this tripod is extremely portable and solid. Ideal for mirrorless cameras or medium format DSLRs.

Key features:

  • Side-pull leg angle selector for fast adjustment
  • Legs can be independently set at 3 angles of spread
  • Ball head for quick and easy framing
  • Rubber leg warmer (pads for camera tripod legs) for comfortable grip
  • Easy link socket for arms and other accessories
  • Fast and secure twist lock system
  • Counterweight hook for extra stability and strap attachment
      Set Up

Unfold the legs as shown in figure 1.
Open the 3 tripod legs. To adjust the tripod's height, release the telescopic extensions on each leg by rotating Twistlock "A" about a quarter of a turn.

You should open the proximal (top) twist‑lock first, then the distal (lower) sections, when extending the legs.

      Leg Angle Adjustment

Each leg can be set independently at any of the 3 angles of spread. To change the angle on a leg, close the leg slightly towards the center column and press down the locking button "W" at the top of the leg. While pressing the button, select the new leg angle and then release button "W" to lock in position, then open the leg fully again. The angle of each leg can be adjusted independently of the other two legs.

Warning: During this procedure, please take care not to insert your fingers inside the central aperture on the spider collar.

     Adjusting Center Column Height

To release the center column "C", unlock gear "D" and adjust the height of the column as required. Tighten gear "D" to lock the column in position.

     Fitting the Quick Release Plate to the Camera

Fix plate "G" to the base of the camera by tightening camera screw in the camera's threaded tripod hole using the ring "Q" WITHOUT APPLYING FORCE. Before fully locking, align the plate "G" with the camera lens. Please ensure you have securely fastened plate "G" to the camera before use. Once fastened, push ring "Q" down so that it lies flat against the plate "G".

    Mounting the Camera on the Head

To mount the camera on the head, open lever "H" while pushing safety catch "X" down, holding them both in the open position whilst attaching the camera by slotting camera plate "G" into the top of the head as shown in figure 5. Release lever "H" and safety catch "X".

Warning: Make sure that the camera is securely locked to the head by pushing lever "H" against plate "G" (Figure 6) and checking that the camera does not move in the head.
From my practical usage experience, lever "H" cannot be completely pushed against plate "G". The reason is the indicated orientation (arrow towards lens) of the plate "G" is reversed.

    Use

The head has independent panoramic and tilt movement.
The lever "F" locks the panoramic movement.
To release the movement, unlock lever "F" by rotating anticlockwise.
Once the desired position is achieved, lock the head by turning lever "F" fully clockwise.

Warning: The lever "P" locks the ball's movement. To ensure safety when using the ball head, always hold the camera with one hand while releasing the ball.

To release the ball "L" when positioning the camera, unlock lever "P" by rotating anticlockwise.
Once the desired position has been reached, lock the ball "L" by turning lever "P" by rotating fully clockwise.
The head features independent fiction adjustment on the ball movement, which makes it easier to position the head before locking it. To adjust the fiction, ensure the ball "L" is not locked. Then hold the camera in one hand and rotate knob "N" clockwise to increase friction. Friction does not lock the camera: we recommended locking the ball "L" by turning the lever "P" fully.

Tripod Manual
Microsoft Copilot
Perplexity AI

2026年5月20日星期三

Software for chip

Bill Gates and Paul Allen kick-started the personal computer revolution by writing Altair BASIC in 1975 to run on the Altair 8800. This software was the first product of their newly formed company, Microsoft.

Stephen Gary Wozniak is an American technology entrepreneur, electrical engineer, computer programmer, and inventor. In 1976, he co-founded Apple Computer (now Apple Inc.) with his early business partner Steve Jobs. Through his work at Apple in the 1970s and 1980s, he is widely recognized as one of the most prominent pioneers of the personal computer revolution.

The Homebrew Computer Club was an early computer hobbyist group in Menlo Park, California, which met from March 1975 to December 1986.

Monte Davidoff is an American computer programmer who was one of the first employees of Microsoft.

couch: a long piece of furniture like a bed, especially in a doctor’s office

rug: a piece of thick material like a small carpet that is used for covering or decorating part of a floor

Traf-O-Data was a business partnership founded in 1972 by Bill Gates, Paul Allen, and Paul Gilbert. The company aimed to read raw data from roadway traffic counters and process it into usable reports for traffic engineers.

reams: reams [plural] (informal) a large quantity of writing

dividend: ​an amount of the profits that a company pays to people who own shares in the company

realm: an area of activity, interest or knowledge

exclusive: only to be used by one particular person or group; only given to one particular person or group

royalty: a sum of money that is paid to somebody who has written a book, piece of music, etc. each time that it is sold or performed

cap at: set as the upper limit

sublicense: a license granted to a third party by a licensee, extending some rights or privileges that the licensee enjoys

viable: feasible; that can be done; that will be successful

ruminate: to think deeply about something

confide: to tell somebody secrets and personal information that you do not want other people to know

goof around: (especially North American English, informal) to spend your time doing silly or stupid things

Bill Gates "Source Code"

Online Dictionaries Used:

hk.dictionary.search.yahoo.com

www.oxfordlearnersdictionaries.com

Google AI Overview

en.wikipedia.org

2026年5月19日星期二

Histogram in Photography

A photography histogram is a graphical representation of an image's tonal values. It displays how pixels are distributed across brightness levels, from pure black on the far left to pure white on the far right. It is an essential tool for evaluating exposure and preventing overexposure or underexposure.

How to Read the Histogram

The horizontal axis (x-axis) represents the brightness of pixels, while the vertical axis (y-axis) represents the number of pixels at that specific brightness.

Left Side (Shadows): Represents pure blacks and dark tones. A spike here means you have dark shadows, but if the graph hits a wall on the far left, you are "crushing" the blacks and losing shadow details.

Middle (Midtones): Represents the middle grays and average light levels in your scene.

Right Side (Highlights): Represents pure whites and bright areas. If the graph hits the right-side wall, you are "clipping" or "blowing out" your highlights (e.g., turning a bright sky into a solid, unrecoverable white).

Common Histogram Shapes

Different types of scenes produce distinct histogram profiles:

High-Key Scene: A bright, airy scene (like a snowfield) will have a histogram heavily weighted toward the right.

Low-Key Scene: A dark, moody scene (like a night cityscape) will have the histogram weighted toward the left.

Balanced Scene: A scene with a good mix of light and shadows will have a bell-curve shape, ideally keeping all data contained within the boundaries.

Google AI overview

List comprehensions

List comprehensions in Python are a concise way to build new lists by looping through an iterable, applying an expression, and optionally filtering with conditions—all in a single line. They are faster and more readable than traditional for loops for simple transformations.

nums = [1, 2, 3, 4]

squares = [n**2 for n in nums]

print(squares)   # [1, 4, 9, 16]


nums = [1, 2, 3, 4, 5, 6]

evens = [n for n in nums if n % 2 == 0]

print(evens)   # [2, 4, 6]

>>> myList = list(range(100))
... filteredList = [item for item in myList if item % 10 == 0]
... print(filteredList)
...
[0, 10, 20, 30, 40, 50, 60, 70, 80, 90]

>>> hisString = 'His name is Samson. He lives in Tuen Mun'
>>> hisString.split('.')
['His name is Samson', ' He lives in Tuen Mun']

>>> hisString.split()
['His', 'name', 'is', 'Samson.', 'He', 'lives', 'in', 'Tuen', 'Mun']

>>> def cleanWord(word):
...         return word.replace('.', '').lower()
...    [cleanWord(word) for word in hisString.split()]
...
['his', 'name', 'is', 'samson', 'he', 'lives', 'in', 'tuen', 'mun']

Outer square brackets [...]

Define that the result will be a list.

Without them, you’d have a generator expression instead (which doesn’t immediately build a list).

.lower()
Converts the string to lowercase.
Ensures consistency (e.g., "Samson" → "samson").

>>> [cleanWord(word) for word in hisString.split() if len(cleanWord(word)) < 3]

['is', 'he', 'in']

word → is a string (e.g., "he", "is", "mun").

You cannot directly compare a string to a number (word < 3) because Python doesn’t know how to order a string against an integer.

Doing so would raise a TypeError.

len(cleanWord(word)) → is an integer (the number of characters in the cleaned word).

This can be compared to another integer (< 3) safely.

Example: "he" → len("he") = 2 → 2 < 3 ✅


Nested list comprehension

>>> [[cleanWord(word) for word in sentence.split()] for sentence in hisString.split('.')]
[['his', 'name', 'is', 'samson'], ['he', 'lives', 'in', 'tuen', 'mun']]

Outer comprehension:
[ ... for sentence in hisString.split('.')]
Splits hisString into sentences using "." as the delimiter (定界符).
Iterates over each sentence.
For each sentence, it produces a list of cleaned words.

Inner comprehension:
[cleanWord(word) for word in sentence.split()]
Splits the current sentence into words (default split on whitespace).
Applies cleanWord(word) to each word.
Produces a list of cleaned words for that sentence.

Inner brackets → build a list of cleaned words per sentence.
Outer brackets → collect those lists into one big list.
Together → you get a nested list comprehension that structures text into sentences → words → cleaned words.

The variable sentence is a string (str).

Microsoft Copilot