Yaohong

Comparison Of Programing Language

Fri, 28 Jun 2024 06:04:20 +0800

Comparison Of Programing Language

Here are the release dates of the following programming languages:

C: C was first released on March 8, 1972.
Objective-C: Objective-C was first released on 1984.
C++: C++ was first released in 1985.
Python: Python was first released on February 20, 1991.
Java: Java was first released on May 23, 1995.
PHP: PHP was first released on June 8, 1995.
JavaScript: PHP was first released on December 4, 1995.

Constraint, Layout An Size On Flutter, Android, iOS And H5

Wed, 26 Jun 2024 09:47:20 +0800

In Flutter, the layout principle of “Constraints go down, Sizes go up, Parent sets position” is a clear and distinct mechanism for handling the layout of widgets. When comparing this with the layout mechanisms in Android, iOS, and HTML5/CSS/JS, we find different approaches and philosophies. Let’s compare these systems to understand their similarities and differences.

Layout and Size in Different Systems

Flutter

Constraints Go Down: Parent widgets send constraints to their children, specifying permissible sizes.
Sizes Go Up: Children choose their size within those constraints.
Parent Sets Position: Parents position children based on their size and constraints.

Android (View System)

Measure Phase:
- Parent Determines Constraints: Parent views provide measurement specifications (MeasureSpec) to children.
- Children Measure Size: Children determine their desired size based on these specifications, including modes (EXACTLY, AT_MOST, UNSPECIFIED).
Layout Phase:
- Parent Sets Size and Position: Parents decide the final size and position of each child view using the layout() method, setting exact bounds within the parent’s coordinate system.

iOS (UIKit)

Constraints (Auto Layout):
- Constraints Propagation: Constraints define relationships between views, such as size ratios or alignments.
- Size Determination: Each view calculates its size based on these constraints and intrinsic content size.
Layout Passes:
- Parent Resolves Constraints: Constraints are resolved to compute the frames (positions and sizes) of views.
- Frames Set Position: The final position and size of each view are determined by resolving the constraints to specific frames.

HTML5/CSS/JS

CSS Box Model:
- Constraints Through CSS: CSS rules (like width, height, max-width, min-width, etc.) define size constraints.
- Size Computation: Elements calculate their size based on CSS rules, content size, and parent constraints.
Layout Mechanisms:
- Flow Layout: Default document flow where elements take up available space in order.
- Flexbox: Parent container defines constraints and alignment for flex items, which then size themselves within these constraints.
- Grid Layout: Parent defines a grid structure, and child elements size and position themselves within grid areas.
Positioning:
- CSS Positioning: Elements can be positioned using properties like position, top, left, right, and bottom, relative to their parent or document flow.

Comparison: Key Aspects

Constraints Handling

Flutter: Constraints explicitly passed from parent to child in a hierarchical manner.
Android: Constraints come from measure specs which include modes and sizes, less explicit than Flutter.
iOS: Constraints are set using Auto Layout, forming a constraint system that is resolved for size and position.
HTML/CSS: Constraints defined using CSS properties; more flexible and varied ways to specify constraints.

Size Determination

Flutter: Children determine size within provided constraints.
Android: Children measure themselves based on measure specs and report back desired sizes.
iOS: Size determined by solving constraints in Auto Layout system.
HTML/CSS: Size determined by CSS rules, intrinsic content, and parent constraints.

Positioning

Flutter: Parent sets position of children after size determination.
Android: Parent uses layout() method to set position within bounds.
iOS: Position determined by resolving constraints to frames.
HTML/CSS: Position determined by CSS properties and layout rules (flow, flexbox, grid).

Flexibility

Flutter: Clear and flexible constraint-based system.
Android: More rigid with fixed measure/layout phases.
iOS: Flexible but complex due to constraint-solving.
HTML/CSS: Highly flexible with multiple layout models (flow, flexbox, grid).

Examples

Flutter Example:

Container(
 constraints: BoxConstraints(
 minWidth: 100,
 maxWidth: 200,
 ),
 child: Text('Hello'),
)

Constraints: Container sends constraints to Text.
Size: Text chooses a size within constraints.
Position: Container sets the position of Text.

Android Example:

<LinearLayout
 android:layout_width="match_parent"
 android:layout_height="match_parent">
 <TextView
 android:layout_width="wrap_content"
 android:layout_height="wrap_content"
 android:text="Hello"/>
</LinearLayout>

Constraints: LinearLayout provides measure specs to TextView.
Size: TextView measures itself based on specs.
Position: LinearLayout sets TextView’s position using layout().

iOS Example:

let container = UIView()
container.translatesAutoresizingMaskIntoConstraints = false

let label = UILabel()
label.translatesAutoresizingMaskIntoConstraints = false
label.text = "Hello"

container.addSubview(label)
NSLayoutConstraint.activate([
 label.widthAnchor.constraint(greaterThanOrEqualToConstant: 100),
 label.widthAnchor.constraint(lessThanOrEqualToConstant: 200),
 label.centerXAnchor.constraint(equalTo: container.centerXAnchor),
 label.centerYAnchor.constraint(equalTo: container.centerYAnchor)
])

Constraints: NSLayoutConstraint sets constraints on label.
Size: UILabel sizes itself based on constraints.
Position: Constraints determine UILabel’s position within container.

HTML/CSS Example:

<div style="display: flex; justify-content: center; align-items: center; width: 200px; height: 200px;">
 <div style="min-width: 100px; max-width: 200px;">
 Hello
 </div>
</div>

Constraints: CSS properties define constraints.
Size: Child div sizes itself based on min and max width.
Position: Parent div positions the child using flexbox.

Summary

The layout systems in Flutter, Android, iOS, and HTML5/CSS/JS each have their own approaches:

What is the purpose of layering the architecture in a Flutter project? How should it be structured?

Thu, 30 May 2024 18:17:20 +0800

What is the purpose of layering the architecture in a Flutter project? How should it be structured?

1.The purpose of layering the architecture in a Flutter project

The purpose of layering the architecture in a Flutter project is to enhance code maintainability, promote a clear separation of concerns, promote reusability, improve testability, and enhance scalability.

2.What is maintainability?

Maintainability is defined as the probability that a failed component or system will be restored or repaired to a specified condition within a specified period or time when maintenance is performed in accordance with prescribed procedures.

Must-know information about Swift

Sat, 25 May 2024 19:55:20 +0800

Must-know information about Swift

entry point

main?

print log

print("Hello swift")
let name = "swift"
print("Hello, \(name)")
print("Hello, \(name1 ?? name)")// If the optional value is missing, the default value is used instead.

Variable, Constant

Int:

On a 32-bit platform, Int is the same size as Int32.
On a 64-bit platform, Int is the same size as Int64.

UInt:

On a 32-bit platform, UInt is the same size as UInt32.
On a 64-bit platform, UInt is the same size as UInt64.

Double represents a 64-bit floating-point number.

The Flutter Plugin Project Files

Fri, 24 May 2024 09:17:20 +0800

The Flutter Plugin Project Files

Creating the plugin project

$ flutter create --org com.example --template=plugin --platforms=android,ios,linux,macos,windows hello_plugin 

Signing iOS app for device deployment using developer identity: "Apple
Development: Yaohong Huang (WS3UUERGF9)"
Creating project hello_plugin...
Resolving dependencies in hello_plugin... (3.3s)
Got dependencies in hello_plugin.
Resolving dependencies in hello_plugin/example... (1.1s)
Got dependencies in hello_plugin/example.
Wrote 161 files.

All done!

Your plugin code is in hello_plugin/lib/hello_plugin.dart.

Your example app code is in hello_plugin/example/lib/main.dart.


Host platform code is in the android, ios, linux, macos, windows directories
under hello_plugin.
To edit platform code in an IDE see
https://flutter.dev/developing-packages/#edit-plugin-package.


To add platforms, run `flutter create -t plugin --platforms <platforms> .` under
hello_plugin.
For more information, see https://flutter.dev/go/plugin-platforms.

hello_plugin
├── CHANGELOG.md
├── LICENSE
├── README.md
├── analysis_options.yaml
├── android
│   ├── build.gradle
│   ├── hello_plugin_android.iml
│   ├── local.properties
│   ├── settings.gradle
│   └── src
│   ├── main
│   └── test
├── example
│   ├── README.md
│   ├── analysis_options.yaml
│   ├── android
│   │   ├── app
│   │   ├── build.gradle
│   │   ├── gradle
│   │   ├── gradle.properties
│   │   ├── gradlew
│   │   ├── gradlew.bat
│   │   ├── hello_plugin_example_android.iml
│   │   ├── local.properties
│   │   └── settings.gradle
│   ├── build
│   │   ├── 52f352800ae67b257006addf8ca3668f
│   │   ├── e29aedf18e567df5728f6b7ce57d2e97.cache.dill.track.dill
│   │   ├── ios
│   │   └── last_build_run.json
│   ├── hello_plugin_example.iml
│   ├── integration_test
│   │   └── plugin_integration_test.dart
│   ├── ios
│   │   ├── Flutter
│   │   ├── Podfile
│   │   ├── Podfile.lock
│   │   ├── Pods
│   │   ├── Runner
│   │   ├── Runner.xcodeproj
│   │   ├── Runner.xcworkspace
│   │   └── RunnerTests
│   ├── lib
│   │   └── main.dart
│   ├── linux
│   │   ├── CMakeLists.txt
│   │   ├── flutter
│   │   ├── main.cc
│   │   ├── my_application.cc
│   │   └── my_application.h
│   ├── macos
│   │   ├── Flutter
│   │   ├── Podfile
│   │   ├── Runner
│   │   ├── Runner.xcodeproj
│   │   ├── Runner.xcworkspace
│   │   └── RunnerTests
│   ├── pubspec.lock
│   ├── pubspec.yaml
│   ├── test
│   │   └── widget_test.dart
│   └── windows
│   ├── CMakeLists.txt
│   ├── flutter
│   └── runner
├── hello_plugin.iml
├── ios
│   ├── Assets
│   ├── Classes
│   │   └── HelloPlugin.swift
│   └── hello_plugin.podspec
├── lib
│   ├── hello_plugin.dart
│   ├── hello_plugin_method_channel.dart
│   └── hello_plugin_platform_interface.dart
├── linux
│   ├── CMakeLists.txt
│   ├── hello_plugin.cc
│   ├── hello_plugin_private.h
│   ├── include
│   │   └── hello_plugin
│   └── test
│   └── hello_plugin_test.cc
├── macos
│   ├── Classes
│   │   └── HelloPlugin.swift
│   └── hello_plugin.podspec
├── pubspec.lock
├── pubspec.yaml
├── test
│   ├── hello_plugin_method_channel_test.dart
│   └── hello_plugin_test.dart
└── windows
 ├── CMakeLists.txt
 ├── hello_plugin.cpp
 ├── hello_plugin.h
 ├── hello_plugin_c_api.cpp
 ├── include
 │   └── hello_plugin
 └── test
 └── hello_plugin_test.cpp

name: hello_plugin
description: A new Flutter plugin project.
version: 0.0.1
homepage:

environment:
 sdk: '>=3.0.5 <4.0.0'
 flutter: ">=3.3.0"

dependencies:
 flutter:
 sdk: flutter
 plugin_platform_interface: ^2.0.2

dev_dependencies:
 flutter_test:
 sdk: flutter
 flutter_lints: ^2.0.0

# For information on the generic Dart part of this file, see the
# following page: https://dart.dev/tools/pub/pubspec

# The following section is specific to Flutter packages.
flutter:
 # This section identifies this Flutter project as a plugin project.
 # The 'pluginClass' specifies the class (in Java, Kotlin, Swift, Objective-C, etc.)
 # which should be registered in the plugin registry. This is required for
 # using method channels.
 # The Android 'package' specifies package in which the registered class is.
 # This is required for using method channels on Android.
 # The 'ffiPlugin' specifies that native code should be built and bundled.
 # This is required for using `dart:ffi`.
 # All these are used by the tooling to maintain consistency when
 # adding or updating assets for this project.
 plugin:
 platforms:
 android:
 package: com.example.hello_plugin
 pluginClass: HelloPlugin
 ios:
 pluginClass: HelloPlugin
 linux:
 pluginClass: HelloPlugin
 macos:
 pluginClass: HelloPlugin
 windows:
 pluginClass: HelloPluginCApi

The Widget tree, Element tree, and Rendering Object tree in flutter

Thu, 23 May 2024 22:17:20 +0800

The purpose of the three type of trees in Flutter:

Widget tree:
- 1.hold the widget config;
- 2.offer a public API;
Element tree:
- 1.manage the lifecycle of widget;
- 2.hold a spot in the UI hierarchy;
- 3.manage parent/child relationship;
Rendering object tree:
- 1.layout, size and paint itself,
- 2.layout children;
- 3.claim input event;

Entry-point binding.dart

void runApp(Widget app){
 WidgetsFlutterBinding.ensureInitialized()
 ..attachRootWidget(app)
 ..scheduleWarmUpFrame();
}

Resouce: https://www.youtube.com/watch?v=996ZgFRENMs

The project files generated by 'flutter create --template=plugin_ffi' using dart:ffi to call C APIs

Thu, 23 May 2024 20:17:20 +0800

The project files generated by ‘flutter create –template=plugin_ffi’ using dart:ffi to call C APIs

Flutter mobile and desktop apps can use the dart:ffi library to call native C APIs.

Here are the project files generated by executing the following command:

flutter create --platforms=android,ios,macos,windows,linux --template=plugin_ffi native_add

flutter version: flutter_macos_3.10.5-stable

Reference:

Notes:

1.The FFI library can only bind against C symbols, so in C++ these symbols are marked extern “C”.

The method with the same name from the last mixin will override the previous ones in dart

Mon, 20 May 2024 18:17:20 +0800

The method with the same name from the last mixin will override the previous ones in dart

TestMixin file:

class MyObject{
 init(){
 print("draw object");
 }
}

mixin Circle{
 
 init(){
 print("draw circle");
 }
}


mixin Square{
 init(){
 print("draw Square");
 }
}

class MyShape extends MyObject with Circle, Square{
 init() {
 super.init();
 print("draw MyShape");
 }
}


void main() {
 MyShape shape = MyShape();
 shape.init();
}

//Output:

draw Square
draw MyShape

Rhythmically Intermittent When my Iphone 12 Connect To MacOS

Tue, 12 Mar 2024 19:01:20 +0800

Rhythmically Intermittent When my Iphone 12 Connect To MacOS

Recently, when my iPhone connects to MacOS, the battery icon on the phone alternates between showing the charging (lightning bolt icon) for a few seconds and then disappearing (no charging battery icon). This continuous loop prevents the phone from charging through MacOS and also hinders the ability to debug iPhone apps.

Mac system version with the issue: macOS Monterey Version 12.7.3 iPhone version with the issue: iOS 16.1

Xcode or Android Studio is unable to list my iPhone device on MacOS

Tue, 21 Mar 2023 14:02:20 +0800

Xcode or Android Studio is unable to list my iPhone device

I encountered an issue where my iPhone12 device was not listed in Android Studio while I was developing.

The issue was dispeared after I restart my macOS, but it occurred again when I reopened my MacBook Pro from sleep mode.

I noticed a new process running in Activity Monitor during the issue occurred.

Here is the process list while android Studio is able to detect my iPhone device:

Distribution failed with errors When developing ios App

Thu, 23 Feb 2023 09:47:20 +0800

Distribution failed with errors When developing ios App

Distribution failed with errors:

Asset validation failed

The product archive is invalid. The Info.plist must contain a LSApplicationCategoryType key, whose value is the UTI for a valid category. For more details, see "Submitting your Mac apps to the App Store". (ID: 67f59c1b-bb08-4694-978f-11d07ff31357)

Solve this issue by add following codes on <project root folder>/macos/Runner/Info.plist:

 <key>LSApplicationCategoryType</key>
 <string>public.app-category.productivity</string>

the value of LSApplicationCategoryType refer https://developer.apple.com/documentation/bundleresources/information_property_list/lsapplicationcategorytype

Flutter HTTP host maven.google.com is not reachable in Windows

Wed, 20 Apr 2022 19:25:20 +0800

Flutter HTTP host `https://maven.google.com/` is not reachable in Windows 10

It shows the following error messages after executing flutter doctor in terminal prompt in Windows 10.

Doctor summary (to see all details, run flutter doctor -v):
[√] Flutter (Channel stable, 2.10.3, on Microsoft Windows [Version 10.0.19044.1586], locale zh-CN)
[√] Android toolchain - develop for Android devices (Android SDK version 30.0.3)
[√] Chrome - develop for the web
[√] Visual Studio - develop for Windows (Visual Studio Community 2019 16.11.2)
[√] Android Studio (version 4.1)
[√] Connected device (4 available)
[!] HTTP Host Availability
 X HTTP host https://maven.google.com/ is not reachable. Reason: An error occurred while checking the HTTP host:
 信号灯超时时间已到

 X HTTP host https://cloud.google.com/ is not reachable. Reason: An error occurred while checking the HTTP host:
 信号灯超时时间已到

How to handle `HTTP host https://maven.google.com/ is not reachable`?

1.Get the port of your http proxy

For example, my https proxy port is 10809, I get it in Option Settings in the v2 proxy. V2 socks port is 10808, https port is sock port +1 which is 10809;

How backward and step are associated with model paramters update?

Wed, 17 Nov 2021 20:27:20 +0800

How are backward and step associated with model paramters update?

optimizer accept the paramters of the model, it can update the parameters, but how is loss function associated with paramters?

loss.backward()

optimizer.step()

REFERENCE:

1.pytorch - connection between loss.backward() and optimizer.step()

2.https://pytorch.org/tutorials/beginner/former_torchies/nnft_tutorial.html#forward-and-backward-function-hooks

nn_Module

Tue, 16 Nov 2021 20:27:20 +0800

nn_Module

1.Where are module parameters configured?

The parameters are stored in the network node which is one of points of a network layer. Neural network layer is defined in init method of module and need to be defined as class variable;

import torch.nn as nn
import numpy as np
class TorchDNN(nn.Module):

 def __init__(self, input, hidden, output):
 super(TorchDNN, self).__init__();
 layer_hidden = nn.Linear(input, hidden, bias = True);
 def forward(self, input_data):
 pass

x = np.array([1, 2, 3])
torch_model = TorchDNN(len(x), 5, 3)
print(torch_model.state_dict())
# OUTPUT:
# OrderedDict()

Network layer should be defined as a variable of Module class;

Github ssh key didn't work

Sun, 05 Sep 2021 14:18:20 +0800

Github ssh key didn’t work

After you config a ssh key on github, you can test your ssh connection.

1.Open Terminal.
2.Enter the following:

$ ssh -T git@github.com

If you see your username of github in the resulting message, that means you configure successfully.

But, when you run git pull under the root of the project, the it prompt you to enter your Username of github, what is wrong?！

Now you should check that the URL in git config file starts with git@, instead of the http or https:

Understanding arange, unsqueeze, repeat, stack methods in Pytorch

Fri, 30 Jul 2021 20:26:20 +0800

Understanding arange, unsqueeze, repeat, stack methods in Pytorch

torch.arange(start=0, end, step=1) return 1-D tensor of size (end-start)/step which value begin from start and each value take with common differences step.
torch.unsqueeze(input, dim) return a new tensor with a dimension of size one insterted at specified position; A dim value within the range [-input.dim() - 1, input.dim() + 1) can be used.
tensor.repeat(size*) return a tensor; the new shape of tensor is that original shape multiplied by arguments correspondingly, if the number of paramter don’t match the original shape, then last dimension of new shape = the last dimension of original shape * last paramter;

L1 L2 Regularization - Optimizer

Mon, 12 Jul 2021 20:26:20 +0800

Optimizer: L1 L2 Regularization

L1,L2 Loss function mean different type of loss function.

L1: sum(Y-f(x)) lasso
L2: sum(Y-f(x))^2 Ridge

L1, L2 regularization :

Y_predict = E(w_i(x_i)+b_i)

MES = E(Y-Y_predict)^2

L1: loss = MSE + 入E|w_i|
L2: loss = MES + 入E(w_i)^2

What does penalize the weights?

It means add another parameters to the loss function, so that the greater the weight, the higher the loss function value. That makes the weight parameters to be less or smaller.

How to Label Voice with Praat for Machine Learning

Sat, 10 Jul 2021 11:34:20 +0800

How to Label Voice with Praat for Machine Learning

1.Install

1.1 Download praat

1.Open Praat: doing Phonetics by Computer website;

2.Choose your OS system on download area in the upper left conner of website;

3.Then click the praat6150_mac.dmg or praat6150_win64.zip to download file;

For example, my os is MacOS, in my case I should download praat6150_mac.dmg and install it.

Option: You can also download the file from github, referce to Praat in github

1.2 Install Phonetic symbols

If you want to see good-quality phonetic characters on your screen and in your clipboard, you have to install the Charis SIL and/or the Doulos SIL font.

Github API Basic Authentication Example

Wed, 07 Jul 2021 20:56:20 +0800

Github API Basic Authentication Example

1.Generate Personal access tokens

Open this page Generate Personal access tokens and click Generate new token to get a token;

2.Use access token to request Github REST api

2.1 Install requests with pip;

pip install requests

2.2 Sustitute GITHUB_API_USER_NAME with your user name and GITHUB_API_PERSONAL_TOKEN with the token you got in step one, then run the following code;

from requests import Request, Session
from requests.exceptions import ConnectionError, Timeout, TooManyRedirects

def getRateLimit():
 url = 'https://api.github.com/rate_limit'
 print(url)
 parameters = {
 }
 headers = {
 'Accept':'application/vnd.github.v3+json',
 }
 session = Session()
 session.auth = ("GITHUB_API_USER_NAME", "GITHUB_API_PERSONAL_TOKEN")
 session.headers.update(headers)
 data = None;
 try:
 response = session.get(url, params=parameters)
 return response;
 except (ConnectionError, Timeout, TooManyRedirects) as e:
 print(e)
 return data;

if __name__ == '__main__':
 rs = getRateLimit()
 # print(rs.headers)
 print(rs.text)


## Output:
## if your access token is correct, the limit of core of resources should be 5000 rather than 60.
{"resources":{"core":{"limit":5000,"used":1245,"remaining":3755,"reset":1625664206},"search":{"limit":30,"used":0,"remaining":30,"reset":1625662728},"graphql":{"limit":5000,"used":0,"remaining":5000,"reset":1625666268},"integration_manifest":{"limit":5000,"used":0,"remaining":5000,"reset":1625666268},"source_import":{"limit":100,"used":0,"remaining":100,"reset":1625662728},"code_scanning_upload":{"limit":500,"used":0,"remaining":500,"reset":1625666268}},"rate":{"limit":5000,"used":1245,"remaining":3755,"reset":1625664206}}

REFERENCE: 1.Other authentication methods

Anacode simple usage

Sat, 26 Jun 2021 13:52:53 +0800

Anacode simple usage

1.1 Download and install –Mac os

Download file: click to download

Install after download.

Run command in terminal to see your anconda version:

$conda -V
conda 4.10.1

Use conda info to see conda configuration:

(base) $ conda info

2.Anaconda Usage

2.1 List all enviroments

(base) $ conda info -e
# conda environments:
#
base /Users/Rhys/opt/anaconda3

2.1 create an enviroment

(base) $ conda create -n py36 python=3.6

2.2 activate an enviroment

(base) $ conda activate py36
(py36) $

The environment had changed after activating;

Simple AI expert Enhanced Loop

Mon, 21 Jun 2021 21:19:20 +0800

Simple AI expert Enhanced Loop

Habit: Daily plan, weekly plan, month plan, 10 minute reading, Daily self-examination

Loop1: Assumption->design a experiment->do->feedback->conclusion

Loop2: Choose a subject->Weekly Share to my classmates->Feedback and update -> Make another share;

The form of our body in future

Mon, 14 Jun 2021 22:15:20 +0800

The form of our body in future

1.The human body consumes energy in the form of carbohydrate.
2.The human body consumes energy in the form of carbohydrate and electric, part of our body are machine.
3.The human body consumes energy in the form of nuclear in long long future, and we can calculate just like we are a super computer .

The Simple Implement of BatchNorm2D

Thu, 27 May 2021 12:30:20 +0800

The Simple Implement of BatchNorm2D

The first is that instead of whiteningthe features in layer inputs and outputs jointly, we will normalize each scalar feature independently, by making ithave the mean of zero and the variance of 1. For a layer with d-dimensional inputx = (x(1). . . x(d)), we will nor-malize each dimension

1.MyBatchNorm2D

import numpy as np;
class MyBatchNorm2D:

 def __init__(self):
 pass

 def forward(self, x):
 x = np.array(x);
 mean = np.mean(x);
 standard_deviation = np.sqrt(np.var(x) + 1e-05);
 x_norm = (x - mean) / standard_deviation;
 return x_norm;

input = [[[[ 1.1713, -10.7508],
 [-2.0155, -0.5290],
 [-0.2751, 1.0233]],
 [[-1.4446, -0.8337],
 [-1.0429, -0.8856],
 [ 5.3324, 7.6233]]],
 [[[ 2.1079, 1.6039],
 [-0.8938, 1.1655],
 [ 8.0355, -0.4911]],
 [[ 3.6337, 10.3400],
 [-1.5365, 0.7931],
 [ 0.8472, 1.1318]]]];
x = np.array(input);
bn = MyBatchNorm2D();
x_norm = bn.forward(input);
print("x_norm:", x_norm);

print("np.mean: ", np.mean(np.array(input)));
print("np.var: " , np.var(np.array(input)));
print("MyBatchNorm2D np.mean: ", np.mean(np.array(x_norm)));
print("MyBatchNorm2D np.var: " , np.var(np.array(x_norm)));

# OUTPUT:
# x_norm: [[[[ 0.0414345 -2.92181622]
# [-0.75064805 -0.38117689]
# [-0.31806977 0.00464894]]
# [[-0.60875025 -0.4569104 ]
# [-0.50890728 -0.4698102 ]
# [ 1.07568036 1.64508601]]]
# [[[ 0.27422743 0.14895769]
# [-0.47184832 0.0399929 ]
# [ 1.74753876 -0.3717568 ]]
# [[ 0.65346666 2.3203247 ]
# [-0.63159209 -0.05256752]
# [-0.0391209 0.03161673]]]]
# np.mean: 1.0045958333333334
# np.var: 16.18707780123264
# MyBatchNorm2D np.mean: 0.0
# MyBatchNorm2D np.var: 0.9999993822236513

2.Using BatchNorm2d in torch

input = [[[[ 1.1713, -10.7508],
 [-2.0155, -0.5290],
 [-0.2751, 1.0233]],
 [[-1.4446, -0.8337],
 [-1.0429, -0.8856],
 [ 5.3324, 7.6233]]],
 [[[ 2.1079, 1.6039],
 [-0.8938, 1.1655],
 [ 8.0355, -0.4911]],
 [[ 3.6337, 10.3400],
 [-1.5365, 0.7931],
 [ 0.8472, 1.1318]]]];

import torch
import torch.nn as nn
input = torch.tensor(input);
bn = nn.BatchNorm2d(2, momentum=None, affine=False, track_running_stats=None)
x_norm = bn(input)
print("BatchNorm2d new_x:", x_norm);

import numpy as np;
print("BatchNorm2d np.mean: " , np.mean(np.array(x_norm)));
print("BatchNorm2d np.var: " , np.var(np.array(x_norm)));

# OUTPUT:
# BatchNorm2d new_x: tensor([[[[ 0.2864, -2.6606],
# [-0.5013, -0.1339],
# [-0.0711, 0.2498]],
# [[-0.9184, -0.7553],
# [-0.8112, -0.7692],
# [ 0.8903, 1.5017]]],
# [[[ 0.5179, 0.3933],
# [-0.2241, 0.2850],
# [ 1.9831, -0.1245]],
# [[ 0.4369, 2.2267],
# [-0.9429, -0.3212],
# [-0.3067, -0.2308]]]])
# BatchNorm2d np.mean: -9.934108e-09
# BatchNorm2d np.var: 0.99999934

REFERENCE:

1.Torch nn.BatchNorm2d

model(x) vs model.forward(x)

Mon, 24 May 2021 11:00:20 +0800

model(x) vs model.forward(x)

__call__ magic method in nn.Module will invoke forward() method and take care of hooks and states that python allows, so we should use model(x) rather than call model.forward(x) directly.

REFERENCE:

1.Why there are different output between model.forward(input) and model(input)

2.Calling forward function without .forward()

3.torch.nn.module codes

How to extract knowledge?

Fri, 21 May 2021 22:04:20 +0800

How to extract knowledge?

I held the opinion before that to do knowledge extraction can be done only by summarizing, but it is not a good way. In contrast, put the knowledge back to a concrete scenario will work for user understandings.

I used to thought common truths are the most valuable of knowledge, but perhaps we need the knowledge with personal experiences.

Because it contains the problem that the author face and the way how one to think, while summary knowledge is only the result of thinking.

How to face the investment risk?

Fri, 14 May 2021 22:44:20 +0800

How to face the investment risk?

Investment risk is the expectation loss of our investment.

It is difficult to calculate the probability of loss sometimes, but we can assume that the loss have happened, then try to find out the reason of loss.

Before investing, we can try to answer the following two questions:

1.What cause the price of properties fall 50%?

Liquidity risk
Credit risk

2.How will you do if the price fell 50%?

DNN RNN CNN codes

Tue, 04 May 2021 08:27:20 +0800

Simple DNN RNN CNN example codes

1.DNN-Deep neural network


import numpy as np;
class myDNN:

 # 3 * 5 * 2
 def __init__(self, input, hidden, output):
 # hidden random weight 
 # Note: hidden_weight can be the shape of (input,hidden); correspondingly, `self.hidden_out` should equal `np.dot(input_data, self.hidden_weight)` to accord with hidden_weight shape.
 self.hidden_weight = np.random.rand(hidden, input); 

 self.hidden_bias = np.random.rand(hidden);

 # hidden random weight 
 self.output_weight = np.random.rand(output,hidden);
 self.output_bias = np.random.rand(output);

 # 
 def forward(self, input_data):
 self.hidden_out = np.dot(input_data, self.hidden_weight.T) + self.hidden_bias;

 self.output_out = np.dot(self.hidden_out, self.output_weight.T) + self.output_bias;

# Usage:
dnn = myDNN(3,5,3);
print("hidden_weight",dnn.hidden_weight)
print("hidden_bias:",dnn.hidden_bias)
print("output_weight",dnn.output_weight)
print("output_bias",dnn.output_bias)

x = np.array([1, 2, 3]) #inut
dnn.forward(x);
print("output_out",dnn.output_out)

# output:
# hidden_weight [[0.99663996 0.39342568 0.5312192 ]
# [0.0798744 0.50312289 0.86241405]
# [0.17138496 0.6761287 0.70645906]
# [0.61662379 0.69389404 0.16623206]
# [0.71213402 0.30800932 0.64149244]]
# hidden_bias: [0.81517457 0.56115705 0.3089624 0.84450962 0.93530796]
# output_weight [[0.34466034 0.31119367 0.12883636 0.34135026 0.43802589]
# [0.31553914 0.16063241 0.8179255 0.52314575 0.79439618]
# [0.86730239 0.25280671 0.20375421 0.78095429 0.67368635]]
# output_bias [0.5588883 0.98722366 0.21507382]
# output_out [ 6.8078659 11.30086755 11.16252686]

1.1 Use DNN in torch

import torch.nn as nn
class TorchDNN(nn.Module):

 def __init__(self, input, hidden, output):
 super(TorchDNN, self).__init__();
 self.layer_hidden = nn.Linear(input, hidden, bias = True);
 self.layer_output = nn.Linear(hidden, output, bias = True);

 # 
 def forward(self, input_data):
 self.hidden_out = self.layer_hidden(input_data);
 self.output_out = self.layer_output(self.hidden_out);

x = np.array([1, 2, 3])
torch_model = TorchDNN(len(x), 5, 3)
print(torch_model.state_dict())

# OrderedDict([('layer_hidden.weight', 
# tensor([[-0.5216, -0.5690, 0.4181],
# [-0.3142, 0.1489, 0.5071],
# [ 0.0295, 0.3381, 0.4401],
# [-0.4697, 0.0732, -0.0328],
# [ 0.5250, 0.1540, 0.2086]])), 
# ('layer_hidden.bias', tensor([-0.5134, 0.2645, -0.3366, -0.0597, 0.0159])), 
# ('layer_output.weight', 
# tensor([[ 0.2770, -0.3408, -0.3145, -0.3686, 0.1060],
# [ 0.1268, 0.0729, -0.3838, 0.2850, 0.1438],
# [ 0.1645, -0.0497, 0.1029, 0.1088, -0.0536]])), 
# ('layer_output.bias', tensor([ 0.0908, -0.1240, 0.2800]))])

2.RNN-Recurrent neural network


import numpy as np;
class myRNN:

 def __init__(self, input, hidden ):
 # random weight
 self.input_hidden_weight = np.random.randint(-10000,10000,(hidden, input))/10000;
 self.hidden_hidden_weight = np.random.randint(-10000,10000,(hidden))/10000;

 # random bias
 self.input_hidden_bias = np.random.randint(-10000,10000,(hidden))/10000;
 self.hidden_hidden_bias = np.random.randint(-10000,10000,(hidden))/10000;
 # self.input_hidden_bias = np.zeros(hidden);
 # self.hidden_hidden_bias = np.zeros(hidden);
 self.hidden_size = hidden

 
 def forward(self, input_data):
 self.last_hidden_output = np.zeros([self.hidden_size]);
 output = []
 for item in input_data:
 # ht=tanh(W_ih * x_t + b_ih  + W_hh*h_(t−1)+b_hh)
 hidden_cur = np.dot(item, self.input_hidden_weight.T) + self.input_hidden_bias;
 hidden_pre = np.dot(self.last_hidden_output, self.hidden_hidden_weight.T) + self.hidden_hidden_bias;

 hidden_output = np.tanh( hidden_cur + hidden_pre )
 output.append(hidden_output)
 self.last_hidden_output = hidden_output;

 return np.array(output), hidden_output;

# diy_model = myRNN(w_ih, w_hh, hidden_size)
x = np.array([[1, 2, 3], [3, 4, 5], [5, 6, 7]]) 
input_size = 3;
hidden_size = 4;
diy_model = myRNN(input_size,hidden_size)
output, hidden_output = diy_model.forward(x)
print("myRNN process output: ", output)
print("myRNN hidden_output:", hidden_output)

# output:
# myRNN process output: [[-0.62745049 -0.99314575 -0.96754221 -0.9965258 ]
# [-0.99542912 -0.99962783 -0.99965698 -0.99998354]
# [-0.99983032 -0.99992543 -0.9999868 -0.99999971]]
# myRNN hidden_output: [-0.99983032 -0.99992543 -0.9999868 -0.99999971]

2.1 Use RNN in torch

import torch.nn as nn;
import torch;
import numpy as np;
class TorchRNN(nn.Module):

 def __init__(self, input_size, hidden):
 super(TorchRNN,self).__init__();
 self.layer = nn.RNN(input_size, hidden, batch_first=True);

 def forward(self, x):
 return self.layer(x)


torch_model = TorchRNN(3, 4)
print(torch_model.state_dict())

x = np.array([[1, 2, 3], [3, 4, 5], [5, 6, 7]]) 
torch_x = torch.FloatTensor([x])
output, h = torch_model.forward(torch_x)
print("output:", output.detach().numpy())
print("h:",h.detach().numpy())

# output: 
# OrderedDict([('layer.weight_ih_l0', tensor([[ 0.0922, 0.2786, -0.4514],
# [ 0.3809, 0.2628, -0.4460],
# [-0.4951, -0.3599, -0.4961],
# [ 0.3794, 0.3397, 0.3185]])), ('layer.weight_hh_l0', tensor([[-0.1330, -0.1843, -0.2618, 0.4246],
# [ 0.4154, -0.3578, -0.4181, -0.4291],
# [ 0.3608, -0.2349, 0.4631, 0.4873],
# [ 0.4886, 0.0285, -0.0490, 0.2928]])), ('layer.bias_ih_l0', tensor([-0.1421, -0.3572, -0.2087, -0.0319])), ('layer.bias_hh_l0', tensor([-0.3799, 0.1126, -0.1766, 0.2630]))])
# output: [[[-0.8416229 -0.589054 -0.99585485 0.9778193 ]
# [-0.45533973 -0.3993926 -0.9999862 0.99957436]
# [-0.6221984 0.05736368 -0.99999994 0.9999955 ]]]
# h: [[[-0.6221984 0.05736368 -0.99999994 0.9999955 ]]]

3.CNN-Convolutional neural network

import numpy as np;
class MyCNN:

 # I don't know how do filters work.
 def __init__(self, in_channel, out_channel, kernel_size):
 # random weight
 # (out_channel, in_channel, kernel_size, kernel_size)
 # self.kernel_weight = np.random.randint(-10000,10000,(out_channel, in_channel, kernel_size, kernel_size))/10000;
 self.kernel_weight = np.array([[[[ 0.0106, -0.1561, 0.0984],
 [ 0.1468, 0.1580, -0.1404],
 [ 0.0856, 0.0780, 0.0636]],
 [[-0.1620, 0.2318, 0.0486],
 [-0.2214, -0.2046, 0.1070],
 [ 0.1609, 0.0160, -0.0374]]],
 [[[ 0.1876, -0.2056, 0.1858],
 [-0.1288, 0.0065, -0.0145],
 [-0.1080, 0.1519, 0.0581]],
 [[-0.0749, 0.2289, -0.0890],
 [ 0.0611, 0.0398, -0.1293],
 [ 0.0911, -0.0264, -0.2104]]]]);
 self.in_channel = in_channel;
 self.out_channel = out_channel;
 self.kernel_size = kernel_size;

 # c*h*w
 def forward(self, input_data):
 output = [];
 input_shape = input_data.shape;

 idx_start = np.int(np.floor( (self.kernel_size)/2)) ;
 width = input_shape[1];
 height = input_shape[2];
 in_channel = input_shape[0];

 for o_c in range(self.out_channel):
 piece_of_out_channel = np.zeros(( width-(idx_start*2), height-(idx_start*2) ));
 # print("piece_of_out_channel:", piece_of_out_channel.shape)
 # width
 for idx_height in range(idx_start, height - idx_start): 
 # height
 for idx_width in range(idx_start, width - idx_start ):
 # kernel_shape_input
 kernel_shape_input = input_data[:, idx_height-idx_start: idx_height+idx_start+1, idx_width-idx_start :idx_width+idx_start+1 ];

 out = self.kernel_weight[o_c] * kernel_shape_input;
 out = np.sum(out)

 # assign value
 idx_h = idx_height - idx_start;
 idx_w = idx_width - idx_start
 piece_of_out_channel[idx_h][idx_w] = out;

 output.append(piece_of_out_channel);
 return output;



# x = np.random.randint(0,10000,(2, 6, 6))/100;
# x = random.astype(int)
x = np.array([[[61,93,18,31,2,49]
 ,[12,62,32,60,58,30]
 ,[49,64,38,74,59,29]
 ,[71,34,29,88,59,41]
 ,[91,72,36,94,79,29]
 ,[17,15,86,29,84,53]]

 ,[[31,25,15,16,35,20]
 ,[76,45,82,88,49,99]
 ,[56,46,82,72,26,55]
 ,[7, 86,32,29,82,91]
 ,[76,68,17,50,19,53]
 ,[87,21,58,35,81,46]]
 ]);
# print(x);
print("x.shape:",x.shape)

myCNN = MyCNN(x.shape[0], 2, 3);
print(myCNN.kernel_weight)
output = myCNN.forward(x)
print("myCNN output:",output)

# output:
# x.shape: (2, 6, 6)
# myCNN output: [array([[-2.5093, 9.0098, -0.2033, 28.9 ],
# [ 6.5155, 27.3464, 0.7038, 14.5031],
# [24.2218, 16.1092, 23.2223, 16.9067],
# [29.6749, 2.0986, 16.8128, 45.025 ]]), 
# array([[-14.77 , -4.3335, 5.0665, 3.2378],
# [-28.2207, 18.1968, 11.889 , -27.3557],
# [ -2.748 , 22.5508, 10.6013, -19.0372],
# [ 22.0148, 9.1788, -22.0313, 9.5176]])]

3.1 Use CNN in torch



import torch;
import torch.nn as nn;
class TorchCNN(nn.Module):

 def __init__(self, in_channel, out_channel, kernel_size ):
 super().__init__();
 self.conv2d = nn.Conv2d(in_channel, out_channel, kernel_size, bias=False);

 def forward(self, input_data):
 return self.conv2d(input_data);

in_channel = x.shape[0];
torchcnn = TorchCNN(in_channel, 2, 3);
print("torchcnn weight:", torchcnn.state_dict())
print("torchcnn weight shape:", torchcnn.state_dict()['conv2d.weight'].numpy().shape)
# torchcnn weight shape: (2, 2, 3, 3) => (out_channel, in_channel, kernel_size, kernel_size)
torch_x = torch.FloatTensor([x])
out = torchcnn.forward(torch_x);
print("TorchCNN out: ", out)


# output
# torchcnn weight: OrderedDict([('conv2d.weight', tensor(
# [[[[ 0.0106, -0.1561, 0.0984],
# [ 0.1468, 0.1580, -0.1404],
# [ 0.0856, 0.0780, 0.0636]],
# [[-0.1620, 0.2318, 0.0486],
# [-0.2214, -0.2046, 0.1070],
# [ 0.1609, 0.0160, -0.0374]]],
# [[[ 0.1876, -0.2056, 0.1858],
# [-0.1288, 0.0065, -0.0145],
# [-0.1080, 0.1519, 0.0581]],
# [[-0.0749, 0.2289, -0.0890],
# [ 0.0611, 0.0398, -0.1293],
# [ 0.0911, -0.0264, -0.2104]]]]))])
# TorchCNN out: tensor([[[[ -2.5066, 9.0144, -0.1983, 28.9003],
# [ 6.5160, 27.3456, 0.7094, 14.5056],
# [ 24.2262, 16.1086, 23.2279, 16.9102],
# [ 29.6763, 2.1047, 16.8210, 45.0250]],
# [[-14.7564, -4.3273, 5.0752, 3.2491],
# [-28.2115, 18.1981, 11.8975, -27.3447],
# [ -2.7442, 22.5540, 10.6096, -19.0247],
# [ 22.0166, 9.1837, -22.0241, 9.5211]]]],
# grad_fn=<MkldnnConvolutionBackward>)

Use Opencv stitching_detailed To Stitch Segmentation image

Thu, 29 Apr 2021 18:36:00 +0800

Use Opencv stitching_detailed To Stitch Segmentation image

Environment:

python version 3.7
opencv-python version 4.5.1.48

1.Download stitching_detailed file save it with file name stitching_detailed.py;

2.Run the command;

$python.exe stitching_detailed.py image_1.png image_2.png image_3.png image_4.png image_5.png image_6.png image_7.png image_8.png image_9.png origin.png --features=brisk --matcher=affine

Note that: image 1~9 is Segmentation images and origin.png is a full picture.

REFERENCE:

stitching_detailed

stitching_detailed.py source codes is follow:

"""
Stitching sample (advanced)
===========================
Show how to use Stitcher API from python.
"""

# Python 2/3 compatibility
from __future__ import print_function

import argparse
from collections import OrderedDict

import cv2 as cv
import numpy as np

EXPOS_COMP_CHOICES = OrderedDict()
EXPOS_COMP_CHOICES['gain_blocks'] = cv.detail.ExposureCompensator_GAIN_BLOCKS
EXPOS_COMP_CHOICES['gain'] = cv.detail.ExposureCompensator_GAIN
EXPOS_COMP_CHOICES['channel'] = cv.detail.ExposureCompensator_CHANNELS
EXPOS_COMP_CHOICES['channel_blocks'] = cv.detail.ExposureCompensator_CHANNELS_BLOCKS
EXPOS_COMP_CHOICES['no'] = cv.detail.ExposureCompensator_NO

BA_COST_CHOICES = OrderedDict()
BA_COST_CHOICES['ray'] = cv.detail_BundleAdjusterRay
BA_COST_CHOICES['reproj'] = cv.detail_BundleAdjusterReproj
BA_COST_CHOICES['affine'] = cv.detail_BundleAdjusterAffinePartial
BA_COST_CHOICES['no'] = cv.detail_NoBundleAdjuster

FEATURES_FIND_CHOICES = OrderedDict()
try:
 cv.xfeatures2d_SURF.create() # check if the function can be called
 FEATURES_FIND_CHOICES['surf'] = cv.xfeatures2d_SURF.create
except (AttributeError, cv.error) as e:
 print("SURF not available")
# if SURF not available, ORB is default
FEATURES_FIND_CHOICES['orb'] = cv.ORB.create
try:
 FEATURES_FIND_CHOICES['sift'] = cv.xfeatures2d_SIFT.create
except AttributeError:
 print("SIFT not available")
try:
 FEATURES_FIND_CHOICES['brisk'] = cv.BRISK_create
except AttributeError:
 print("BRISK not available")
try:
 FEATURES_FIND_CHOICES['akaze'] = cv.AKAZE_create
except AttributeError:
 print("AKAZE not available")

SEAM_FIND_CHOICES = OrderedDict()
SEAM_FIND_CHOICES['gc_color'] = cv.detail_GraphCutSeamFinder('COST_COLOR')
SEAM_FIND_CHOICES['gc_colorgrad'] = cv.detail_GraphCutSeamFinder('COST_COLOR_GRAD')
SEAM_FIND_CHOICES['dp_color'] = cv.detail_DpSeamFinder('COLOR')
SEAM_FIND_CHOICES['dp_colorgrad'] = cv.detail_DpSeamFinder('COLOR_GRAD')
SEAM_FIND_CHOICES['voronoi'] = cv.detail.SeamFinder_createDefault(cv.detail.SeamFinder_VORONOI_SEAM)
SEAM_FIND_CHOICES['no'] = cv.detail.SeamFinder_createDefault(cv.detail.SeamFinder_NO)

ESTIMATOR_CHOICES = OrderedDict()
ESTIMATOR_CHOICES['homography'] = cv.detail_HomographyBasedEstimator
ESTIMATOR_CHOICES['affine'] = cv.detail_AffineBasedEstimator

WARP_CHOICES = (
 'spherical',
 'plane',
 'affine',
 'cylindrical',
 'fisheye',
 'stereographic',
 'compressedPlaneA2B1',
 'compressedPlaneA1.5B1',
 'compressedPlanePortraitA2B1',
 'compressedPlanePortraitA1.5B1',
 'paniniA2B1',
 'paniniA1.5B1',
 'paniniPortraitA2B1',
 'paniniPortraitA1.5B1',
 'mercator',
 'transverseMercator',
)

WAVE_CORRECT_CHOICES = OrderedDict()
WAVE_CORRECT_CHOICES['horiz'] = cv.detail.WAVE_CORRECT_HORIZ
WAVE_CORRECT_CHOICES['no'] = None
WAVE_CORRECT_CHOICES['vert'] = cv.detail.WAVE_CORRECT_VERT

BLEND_CHOICES = ('multiband', 'feather', 'no',)

parser = argparse.ArgumentParser(
 prog="stitching_detailed.py", description="Rotation model images stitcher"
)
parser.add_argument(
 'img_names', nargs='+',
 help="Files to stitch", type=str
)
parser.add_argument(
 '--try_cuda',
 action='store',
 default=False,
 help="Try to use CUDA. The default value is no. All default values are for CPU mode.",
 type=bool, dest='try_cuda'
)
parser.add_argument(
 '--work_megapix', action='store', default=0.6,
 help="Resolution for image registration step. The default is 0.6 Mpx",
 type=float, dest='work_megapix'
)
parser.add_argument(
 '--features', action='store', default=list(FEATURES_FIND_CHOICES.keys())[0],
 help="Type of features used for images matching. The default is '%s'." % list(FEATURES_FIND_CHOICES.keys())[0],
 choices=FEATURES_FIND_CHOICES.keys(),
 type=str, dest='features'
)
parser.add_argument(
 '--matcher', action='store', default='homography',
 help="Matcher used for pairwise image matching. The default is 'homography'.",
 choices=('homography', 'affine'),
 type=str, dest='matcher'
)
parser.add_argument(
 '--estimator', action='store', default=list(ESTIMATOR_CHOICES.keys())[0],
 help="Type of estimator used for transformation estimation. The default is '%s'." % list(ESTIMATOR_CHOICES.keys())[0],
 choices=ESTIMATOR_CHOICES.keys(),
 type=str, dest='estimator'
)
parser.add_argument(
 '--match_conf', action='store',
 help="Confidence for feature matching step. The default is 0.3 for ORB and 0.65 for other feature types.",
 type=float, dest='match_conf'
)
parser.add_argument(
 '--conf_thresh', action='store', default=1.0,
 help="Threshold for two images are from the same panorama confidence.The default is 1.0.",
 type=float, dest='conf_thresh'
)
parser.add_argument(
 '--ba', action='store', default=list(BA_COST_CHOICES.keys())[0],
 help="Bundle adjustment cost function. The default is '%s'." % list(BA_COST_CHOICES.keys())[0],
 choices=BA_COST_CHOICES.keys(),
 type=str, dest='ba'
)
parser.add_argument(
 '--ba_refine_mask', action='store', default='xxxxx',
 help="Set refinement mask for bundle adjustment. It looks like 'x_xxx', "
 "where 'x' means refine respective parameter and '_' means don't refine, "
 "and has the following format:<fx><skew><ppx><aspect><ppy>. "
 "The default mask is 'xxxxx'. "
 "If bundle adjustment doesn't support estimation of selected parameter then "
 "the respective flag is ignored.",
 type=str, dest='ba_refine_mask'
)
parser.add_argument(
 '--wave_correct', action='store', default=list(WAVE_CORRECT_CHOICES.keys())[0],
 help="Perform wave effect correction. The default is '%s'" % list(WAVE_CORRECT_CHOICES.keys())[0],
 choices=WAVE_CORRECT_CHOICES.keys(),
 type=str, dest='wave_correct'
)
parser.add_argument(
 '--save_graph', action='store', default=None,
 help="Save matches graph represented in DOT language to <file_name> file.",
 type=str, dest='save_graph'
)
parser.add_argument(
 '--warp', action='store', default=WARP_CHOICES[0],
 help="Warp surface type. The default is '%s'." % WARP_CHOICES[0],
 choices=WARP_CHOICES,
 type=str, dest='warp'
)
parser.add_argument(
 '--seam_megapix', action='store', default=0.1,
 help="Resolution for seam estimation step. The default is 0.1 Mpx.",
 type=float, dest='seam_megapix'
)
parser.add_argument(
 '--seam', action='store', default=list(SEAM_FIND_CHOICES.keys())[0],
 help="Seam estimation method. The default is '%s'." % list(SEAM_FIND_CHOICES.keys())[0],
 choices=SEAM_FIND_CHOICES.keys(),
 type=str, dest='seam'
)
parser.add_argument(
 '--compose_megapix', action='store', default=-1,
 help="Resolution for compositing step. Use -1 for original resolution. The default is -1",
 type=float, dest='compose_megapix'
)
parser.add_argument(
 '--expos_comp', action='store', default=list(EXPOS_COMP_CHOICES.keys())[0],
 help="Exposure compensation method. The default is '%s'." % list(EXPOS_COMP_CHOICES.keys())[0],
 choices=EXPOS_COMP_CHOICES.keys(),
 type=str, dest='expos_comp'
)
parser.add_argument(
 '--expos_comp_nr_feeds', action='store', default=1,
 help="Number of exposure compensation feed.",
 type=np.int32, dest='expos_comp_nr_feeds'
)
parser.add_argument(
 '--expos_comp_nr_filtering', action='store', default=2,
 help="Number of filtering iterations of the exposure compensation gains.",
 type=float, dest='expos_comp_nr_filtering'
)
parser.add_argument(
 '--expos_comp_block_size', action='store', default=32,
 help="BLock size in pixels used by the exposure compensator. The default is 32.",
 type=np.int32, dest='expos_comp_block_size'
)
parser.add_argument(
 '--blend', action='store', default=BLEND_CHOICES[0],
 help="Blending method. The default is '%s'." % BLEND_CHOICES[0],
 choices=BLEND_CHOICES,
 type=str, dest='blend'
)
parser.add_argument(
 '--blend_strength', action='store', default=5,
 help="Blending strength from [0,100] range. The default is 5",
 type=np.int32, dest='blend_strength'
)
parser.add_argument(
 '--output', action='store', default='result.jpg',
 help="The default is 'result.jpg'",
 type=str, dest='output'
)
parser.add_argument(
 '--timelapse', action='store', default=None,
 help="Output warped images separately as frames of a time lapse movie, "
 "with 'fixed_' prepended to input file names.",
 type=str, dest='timelapse'
)
parser.add_argument(
 '--rangewidth', action='store', default=-1,
 help="uses range_width to limit number of images to match with.",
 type=int, dest='rangewidth'
)

__doc__ += '\n' + parser.format_help()


def get_matcher(args):
 try_cuda = args.try_cuda
 matcher_type = args.matcher
 if args.match_conf is None:
 if args.features == 'orb':
 match_conf = 0.3
 else:
 match_conf = 0.65
 else:
 match_conf = args.match_conf
 range_width = args.rangewidth
 if matcher_type == "affine":
 matcher = cv.detail_AffineBestOf2NearestMatcher(False, try_cuda, match_conf)
 elif range_width == -1:
 matcher = cv.detail.BestOf2NearestMatcher_create(try_cuda, match_conf)
 else:
 matcher = cv.detail.BestOf2NearestRangeMatcher_create(range_width, try_cuda, match_conf)
 return matcher


def get_compensator(args):
 expos_comp_type = EXPOS_COMP_CHOICES[args.expos_comp]
 expos_comp_nr_feeds = args.expos_comp_nr_feeds
 expos_comp_block_size = args.expos_comp_block_size
 # expos_comp_nr_filtering = args.expos_comp_nr_filtering
 if expos_comp_type == cv.detail.ExposureCompensator_CHANNELS:
 compensator = cv.detail_ChannelsCompensator(expos_comp_nr_feeds)
 # compensator.setNrGainsFilteringIterations(expos_comp_nr_filtering)
 elif expos_comp_type == cv.detail.ExposureCompensator_CHANNELS_BLOCKS:
 compensator = cv.detail_BlocksChannelsCompensator(
 expos_comp_block_size, expos_comp_block_size,
 expos_comp_nr_feeds
 )
 # compensator.setNrGainsFilteringIterations(expos_comp_nr_filtering)
 else:
 compensator = cv.detail.ExposureCompensator_createDefault(expos_comp_type)
 return compensator


def main():
 args = parser.parse_args()
 img_names = args.img_names
 print(img_names)
 work_megapix = args.work_megapix
 seam_megapix = args.seam_megapix
 compose_megapix = args.compose_megapix
 conf_thresh = args.conf_thresh
 ba_refine_mask = args.ba_refine_mask
 wave_correct = WAVE_CORRECT_CHOICES[args.wave_correct]
 if args.save_graph is None:
 save_graph = False
 else:
 save_graph = True
 warp_type = args.warp
 blend_type = args.blend
 blend_strength = args.blend_strength
 result_name = args.output
 if args.timelapse is not None:
 timelapse = True
 if args.timelapse == "as_is":
 timelapse_type = cv.detail.Timelapser_AS_IS
 elif args.timelapse == "crop":
 timelapse_type = cv.detail.Timelapser_CROP
 else:
 print("Bad timelapse method")
 exit()
 else:
 timelapse = False
 finder = FEATURES_FIND_CHOICES[args.features]()
 seam_work_aspect = 1
 full_img_sizes = []
 features = []
 images = []
 is_work_scale_set = False
 is_seam_scale_set = False
 is_compose_scale_set = False
 for name in img_names:
 full_img = cv.imread(cv.samples.findFile(name))
 if full_img is None:
 print("Cannot read image ", name)
 exit()
 full_img_sizes.append((full_img.shape[1], full_img.shape[0]))
 if work_megapix < 0:
 img = full_img
 work_scale = 1
 is_work_scale_set = True
 else:
 if is_work_scale_set is False:
 work_scale = min(1.0, np.sqrt(work_megapix * 1e6 / (full_img.shape[0] * full_img.shape[1])))
 is_work_scale_set = True
 img = cv.resize(src=full_img, dsize=None, fx=work_scale, fy=work_scale, interpolation=cv.INTER_LINEAR_EXACT)
 if is_seam_scale_set is False:
 seam_scale = min(1.0, np.sqrt(seam_megapix * 1e6 / (full_img.shape[0] * full_img.shape[1])))
 seam_work_aspect = seam_scale / work_scale
 is_seam_scale_set = True
 img_feat = cv.detail.computeImageFeatures2(finder, img)
 features.append(img_feat)
 img = cv.resize(src=full_img, dsize=None, fx=seam_scale, fy=seam_scale, interpolation=cv.INTER_LINEAR_EXACT)
 images.append(img)

 matcher = get_matcher(args)
 p = matcher.apply2(features)
 matcher.collectGarbage()

 if save_graph:
 with open(args.save_graph, 'w') as fh:
 fh.write(cv.detail.matchesGraphAsString(img_names, p, conf_thresh))

 indices = cv.detail.leaveBiggestComponent(features, p, conf_thresh)
 img_subset = []
 img_names_subset = []
 full_img_sizes_subset = []
 for i in range(len(indices)):
 img_names_subset.append(img_names[indices[i, 0]])
 img_subset.append(images[indices[i, 0]])
 full_img_sizes_subset.append(full_img_sizes[indices[i, 0]])
 images = img_subset
 img_names = img_names_subset
 full_img_sizes = full_img_sizes_subset
 num_images = len(img_names)
 if num_images < 2:
 print("Need more images")
 exit()

 estimator = ESTIMATOR_CHOICES[args.estimator]()
 b, cameras = estimator.apply(features, p, None)
 if not b:
 print("Homography estimation failed.")
 exit()
 for cam in cameras:
 cam.R = cam.R.astype(np.float32)

 adjuster = BA_COST_CHOICES[args.ba]()
 adjuster.setConfThresh(1)
 refine_mask = np.zeros((3, 3), np.uint8)
 if ba_refine_mask[0] == 'x':
 refine_mask[0, 0] = 1
 if ba_refine_mask[1] == 'x':
 refine_mask[0, 1] = 1
 if ba_refine_mask[2] == 'x':
 refine_mask[0, 2] = 1
 if ba_refine_mask[3] == 'x':
 refine_mask[1, 1] = 1
 if ba_refine_mask[4] == 'x':
 refine_mask[1, 2] = 1
 adjuster.setRefinementMask(refine_mask)
 b, cameras = adjuster.apply(features, p, cameras)
 if not b:
 print("Camera parameters adjusting failed.")
 exit()
 focals = []
 for cam in cameras:
 focals.append(cam.focal)
 focals.sort()
 if len(focals) % 2 == 1:
 warped_image_scale = focals[len(focals) // 2]
 else:
 warped_image_scale = (focals[len(focals) // 2] + focals[len(focals) // 2 - 1]) / 2
 if wave_correct is not None:
 rmats = []
 for cam in cameras:
 rmats.append(np.copy(cam.R))
 rmats = cv.detail.waveCorrect(rmats, wave_correct)
 for idx, cam in enumerate(cameras):
 cam.R = rmats[idx]
 corners = []
 masks_warped = []
 images_warped = []
 sizes = []
 masks = []
 for i in range(0, num_images):
 um = cv.UMat(255 * np.ones((images[i].shape[0], images[i].shape[1]), np.uint8))
 masks.append(um)

 warper = cv.PyRotationWarper(warp_type, warped_image_scale * seam_work_aspect) # warper could be nullptr?
 for idx in range(0, num_images):
 K = cameras[idx].K().astype(np.float32)
 swa = seam_work_aspect
 K[0, 0] *= swa
 K[0, 2] *= swa
 K[1, 1] *= swa
 K[1, 2] *= swa
 corner, image_wp = warper.warp(images[idx], K, cameras[idx].R, cv.INTER_LINEAR, cv.BORDER_REFLECT)
 corners.append(corner)
 sizes.append((image_wp.shape[1], image_wp.shape[0]))
 images_warped.append(image_wp)
 p, mask_wp = warper.warp(masks[idx], K, cameras[idx].R, cv.INTER_NEAREST, cv.BORDER_CONSTANT)
 masks_warped.append(mask_wp.get())

 images_warped_f = []
 for img in images_warped:
 imgf = img.astype(np.float32)
 images_warped_f.append(imgf)

 compensator = get_compensator(args)
 compensator.feed(corners=corners, images=images_warped, masks=masks_warped)

 seam_finder = SEAM_FIND_CHOICES[args.seam]
 seam_finder.find(images_warped_f, corners, masks_warped)
 compose_scale = 1
 corners = []
 sizes = []
 blender = None
 timelapser = None
 # https://github.com/opencv/opencv/blob/master/samples/cpp/stitching_detailed.cpp#L725 ?
 for idx, name in enumerate(img_names):
 full_img = cv.imread(name)
 if not is_compose_scale_set:
 if compose_megapix > 0:
 compose_scale = min(1.0, np.sqrt(compose_megapix * 1e6 / (full_img.shape[0] * full_img.shape[1])))
 is_compose_scale_set = True
 compose_work_aspect = compose_scale / work_scale
 warped_image_scale *= compose_work_aspect
 warper = cv.PyRotationWarper(warp_type, warped_image_scale)
 for i in range(0, len(img_names)):
 cameras[i].focal *= compose_work_aspect
 cameras[i].ppx *= compose_work_aspect
 cameras[i].ppy *= compose_work_aspect
 sz = (full_img_sizes[i][0] * compose_scale, full_img_sizes[i][1] * compose_scale)
 K = cameras[i].K().astype(np.float32)
 roi = warper.warpRoi(sz, K, cameras[i].R)
 corners.append(roi[0:2])
 sizes.append(roi[2:4])
 if abs(compose_scale - 1) > 1e-1:
 img = cv.resize(src=full_img, dsize=None, fx=compose_scale, fy=compose_scale,
 interpolation=cv.INTER_LINEAR_EXACT)
 else:
 img = full_img
 _img_size = (img.shape[1], img.shape[0])
 K = cameras[idx].K().astype(np.float32)
 corner, image_warped = warper.warp(img, K, cameras[idx].R, cv.INTER_LINEAR, cv.BORDER_REFLECT)
 mask = 255 * np.ones((img.shape[0], img.shape[1]), np.uint8)
 p, mask_warped = warper.warp(mask, K, cameras[idx].R, cv.INTER_NEAREST, cv.BORDER_CONSTANT)
 compensator.apply(idx, corners[idx], image_warped, mask_warped)
 image_warped_s = image_warped.astype(np.int16)
 dilated_mask = cv.dilate(masks_warped[idx], None)
 seam_mask = cv.resize(dilated_mask, (mask_warped.shape[1], mask_warped.shape[0]), 0, 0, cv.INTER_LINEAR_EXACT)
 mask_warped = cv.bitwise_and(seam_mask, mask_warped)
 if blender is None and not timelapse:
 blender = cv.detail.Blender_createDefault(cv.detail.Blender_NO)
 dst_sz = cv.detail.resultRoi(corners=corners, sizes=sizes)
 blend_width = np.sqrt(dst_sz[2] * dst_sz[3]) * blend_strength / 100
 if blend_width < 1:
 blender = cv.detail.Blender_createDefault(cv.detail.Blender_NO)
 elif blend_type == "multiband":
 blender = cv.detail_MultiBandBlender()
 blender.setNumBands((np.log(blend_width) / np.log(2.) - 1.).astype(np.int))
 elif blend_type == "feather":
 blender = cv.detail_FeatherBlender()
 blender.setSharpness(1. / blend_width)
 blender.prepare(dst_sz)
 elif timelapser is None and timelapse:
 timelapser = cv.detail.Timelapser_createDefault(timelapse_type)
 timelapser.initialize(corners, sizes)
 if timelapse:
 ma_tones = np.ones((image_warped_s.shape[0], image_warped_s.shape[1]), np.uint8)
 timelapser.process(image_warped_s, ma_tones, corners[idx])
 pos_s = img_names[idx].rfind("/")
 if pos_s == -1:
 fixed_file_name = "fixed_" + img_names[idx]
 else:
 fixed_file_name = img_names[idx][:pos_s + 1] + "fixed_" + img_names[idx][pos_s + 1:]
 cv.imwrite(fixed_file_name, timelapser.getDst())
 else:
 blender.feed(cv.UMat(image_warped_s), mask_warped, corners[idx])
 if not timelapse:
 result = None
 result_mask = None
 result, result_mask = blender.blend(result, result_mask)
 cv.imwrite(result_name, result)
 zoom_x = 600.0 / result.shape[1]
 dst = cv.normalize(src=result, dst=None, alpha=255., norm_type=cv.NORM_MINMAX, dtype=cv.CV_8U)
 dst = cv.resize(dst, dsize=None, fx=zoom_x, fy=zoom_x)
 cv.imshow(result_name, dst)
 cv.waitKey()

 print("Done")


if __name__ == '__main__':
 print(__doc__)
 main()
 cv.destroyAllWindows()

How to categorize our money

Wed, 28 Apr 2021 23:25:20 +0800

How to categorize our invest money?

I think sleep peacefully at in night is the most important thing when we do the investment. I don’t need to worry about that I can’t pay my next monthly rent for the apartment after investing.

I believe that to avoid wake up at midnight is categorize our money. We can divide our cash into four categories.

The first category is called daily cash, it used to pay our daily cost. Apparently this type of money can not be invested to high risk stocks, or we might wake up at midnight.

2021-04-14 Diary

Wed, 14 Apr 2021 23:20:20 +0800

2021-04-14 Diary

While I was sitting on the chair searching on the internet, my little baby looked at me and was going to cry in the bedroom. I known he was tired and need to go to bed. So I close the Mac and going in the bedroom. When I clapped and open my hands to hold him, he laughed happily and came to me.

I held him to the bed, but as soon as he sit on the bed, he started to cry and crawled slowly to the bedside cupboard. I thought he may want to milk, but his mother was taking a bath. My little baby often milked before he go to bed.

Futures learning notes 001

Wed, 07 Apr 2021 22:40:20 +0800

Futures learning notes 001

Futures

Futures are derivate financial contract that obligate the parties to transact an asset at a predetermined future date and price.

The buyer must purchase or the seller must sell the underlying asset at the set price, regardless of the current market price at the expiration date.

futures can exchange in Shanghai Futures Exchange

Shanghai Futures Exchange can exchange follow futures:

Copper, Copper(BC), Lead, Zinc, Aluminium, Nickel, Tin

从个人角度的行业划分

Sun, 28 Mar 2021 00:48:20 +0800

从个人角度的行业划分

衣食住行

衣（必）：服装、个护
食（必）：农业、餐饮、
住（必）：房产、家电（+联网+智能）、
行：车、物流、公共交通

医师作息

医疗：药品
教育：英语、搜索
工作：会议、办公耗材
休息&娱乐：

Acceptance in meditation

Sun, 21 Mar 2021 19:58:20 +0800

Acceptance in meditation

“Bring your attention to your head and face, noticing all sensations and feeling here. And just trying to accept all the feelings and sensations, rather than trying to judge them or change them in any way.”

This pratice works for me and alleviates my tension headache significantly.

Tension headache

Sun, 14 Mar 2021 21:15:20 +0800

Tension headache

Symptoms

Signs and symptoms of a tension headache include:

Dull, aching head pain
Sensation of tightness or pressure across your forehead or on the sides and back of your head
Tenderness on your scalp, neck and shoulder muscles

Tension headaches are divided into two main categories — episodic and chronic.

Episodic tension headaches

Episodic tension headaches can last from 30 minutes to a week. Frequent episodic tension headaches occur less than 15 days a month for at least three months. Frequent episodic tension headaches may become chronic.

高等数学主要内容

Sun, 07 Mar 2021 20:14:20 +0800

高等数学主要内容

一、空间解析几何与向量代数

空间直角坐标系
向量的概念及其线性；
向量坐标、向量的数量积；
平面方程、直线方程；
曲面方程、曲线方程；
二次曲面；

二、多元函数的微分学

二元函数的极限与连续；
【重点】偏导数；
高阶偏导数；
全微分；
复合函数求导法则；
隐函数求导法则；
空间曲线的切线与法平面、曲面的切平面与法线；
二元函数的极值；

三、重积分

二重积分和三重积分的定义与性质；
二重积分的计算；
三重积分的计算；
重积分的应用：曲面面积、曲顶柱体的体积、平面薄板的质量；

四、曲线积分与曲面积分

对弧长曲线积分，对坐标的曲线积分的性质；
两类曲线积分的计算；
平面曲线积分与路径无关的条件
两类曲面积分的定义与性质
两类曲面积分的计算；

五、常微分方程

微分方程的一般概念
三类一阶微分方程；
三类可降阶的二阶微分方程；
二阶线性微分方程解的结构；
二阶常系数线性齐次微分方程；

六、无穷级数

数项级数的概念与性质；
正项级数及其审敛法；
一般项级数的审敛法；
幂级数的收敛性和函数性质；
函数的仄勒级数展开式；
傅里叶级数；

高等数学主要内容

Sun, 07 Mar 2021 20:14:20 +0800

高等数学主要内容

一、空间解析几何与向量代数

空间直角坐标系
向量的概念及其线性；
向量坐标、向量的数量积；
平面方程、直线方程；
曲面方程、曲线方程；
二次曲面；

二、多元函数的微分学

二元函数的极限与连续；
【重点】偏导数；
高阶偏导数；
全微分；
复合函数求导法则；
隐函数求导法则；
空间曲线的切线与法平面、曲面的切平面与法线；
二元函数的极值；

三、重积分

二重积分和三重积分的定义与性质；
二重积分的计算；
三重积分的计算；
重积分的应用：曲面面积、曲顶柱体的体积、平面薄板的质量；

四、曲线积分与曲面积分

对弧长曲线积分，对坐标的曲线积分的性质；
两类曲线积分的计算；
平面曲线积分与路径无关的条件
两类曲面积分的定义与性质
两类曲面积分的计算；

五、常微分方程

微分方程的一般概念
三类一阶微分方程；
三类可降阶的二阶微分方程；
二阶线性微分方程解的结构；
二阶常系数线性齐次微分方程；

六、无穷级数

数项级数的概念与性质；
正项级数及其审敛法；
一般项级数的审敛法；
幂级数的收敛性和函数性质；
函数的仄勒级数展开式；
傅里叶级数；

Laravel配置阿里云企业邮箱

Sun, 28 Feb 2021 23:15:20 +0800

Laravel配置阿里云企业邮箱

企业邮箱POP、SMTP、IMAP地址列表如下：

（阿里云邮箱web端通用访问地址：https://qiye.aliyun.com/），客户端推荐以下参数配置：

How to persuade people using motivational interviewing?

Sat, 20 Feb 2021 23:19:20 +0800

How to persuade people using motivational interviewing?

Imaging a person that you want to advise him/her to do or not to do something.

I image I was advising my wife to sleep earlier, because she take care our baby at home and often sleep too late.

1. Ask her an open question about the behaviour?

For example, “Why you do that? I would like to hear how to you think that sincerely?”

The past five days

Sun, 14 Feb 2021 19:15:20 +0800

The past five days

I have been working every day for the past five days.

My plan is:

day1: set up the development enviroment and finish login page; # interface1: login [done]
day2: finish home page layout and choose a image from backend; # interface7: getImages; [part done]
day3: load a image and texts; # interface8: getTextList; [done]
day4: drag text by mouse, assign the text size and color; [part done]
day5: finish font style plug in, load the setting text style;
day6: interface: save style interface;

The reality is:

Why do I see it a burden

Sun, 07 Feb 2021 22:29:20 +0800

Why do I see it a burden?

1.The whole story

My friend asked me for a help recently. If I promise him, it would take me at least a week to finish the task which I can spend to improve my skill and prepare for the examination on April.

The task is to build a website to collect data which is important for him because he needs the data for research.

How does warp Perspective work?

Sat, 23 Jan 2021 16:29:20 +0800

How does warp Perspective work?

1.warp perspective with cv2

if __name__ == "__main__":
 # coordinate: (y,x), left_top, right_rop, left_bottom, right_bottom
 src = np.float32([[20.0, 0.0], [20.0 ,315.0], [186.0, 17.2], [181.0, 299.0]])
 dst = np.float32([[0.0, 0.0], [0.0, 315.0], [202.0, 7.0], [200.0, 306.0]])

 # load image
 warp_img = cv2.imread("./my_wide_angle_orig.jpg")
 warp_img = cv2.cvtColor(warp_img, cv2.COLOR_BGR2RGB)
 print("warp_img: ",warp_img.shape) # (638, 958, 3)
 width = int(warp_img.shape[1]/3)
 height = int(warp_img.shape[0]/3)
 warp_img = cv2.resize(warp_img, (width,height), interpolation=cv2.INTER_LINEAR)
 print("warp_img.shape:",warp_img.shape) # (212, 319, 3)

 ## orig image
 plt.subplot(121),
 plt.title("warp_img")
 plt.imshow(warp_img)

 # cv2 warp perspective
 cv2_matrix = cv2.getPerspectiveTransform(src, dst)
 print("cv2_matrix:\n",cv2_matrix)
 cv2_fix_img = cv2.warpPerspective(warp_img, cv2_matrix, (width,height))
 plt.subplot(122),
 plt.title('cv2_fix_img')
 plt.imshow(cv2_fix_img) 
 plt.show()

2.Implement it in our way

Step1 calculate warp matrix:

my_warp_matrix reshape:
 [[ 1.13729359e+00 -8.24289989e-18 -2.27458717e+01]
 [-6.10786436e-02 9.69843448e-01 1.22157287e+00]
 [-3.44743207e-04 -7.38466280e-05 1.00000000e+00]]

Step2. use the warp matrix to warp perspective

codes are:

How to Read a Image in Python

Sat, 23 Jan 2021 11:26:20 +0800

How to Read a Image in Python

There are three ways to read a image, the codes is showing below.


image_path = "fasterRCNN/faster-rcnn-keras-master/img/street.jpg";

from PIL import Image
import numpy as np
image = Image.open(image_path) # RGB
tmp_image = np.array(image)
print("PIL open shape:",tmp_image.shape, tmp_image)


# load with cv2
import cv2
image = cv2.imread(image_path) # mode: BGR
print("cv2 imread shape:", image.shape) # (width,height,channel)


# load with matplotlib
import matplotlib.image as mpimg
image = mpimg.imread(image_path) # mode: RGB
print("matplotlib imread shape:",image.shape, image) # ( height, width,channel(RGB) )

If you use cv2 to load a image , then show it with matplotlib, you should convert the image to RGB mode, this is because cv2 read a image in BGR mode, while matplotlib presents the image in RGB mode;

Differences On Numpyp.Floor And Python Int method.md

Thu, 31 Dec 2020 12:35:20 +0800

Differences On Numpyp.Floor And Python Int method.md

numpy.floor Return the floor of the input.

The floor of the scalar x is the largest integer i, such that i <= x;

np.floor() will not change its data type;

Example:

import numpy as np

tmp_list = list([-3.3, -2.22, -1.56, -0.56, 0.56, 1.56, 2.22, 3.33])

tmp_list = np.array(tmp_list)
print("type(tmp_list[0]):", type(tmp_list[0]) )
print("type(np.floor(tmp_list)[0]):", type(np.floor(tmp_list)[0]) )
print("np.floor(tmp_list):", np.floor(tmp_list))

# Output:
# type(tmp_list[0]): <class 'numpy.float64'>
# type(np.floor(tmp_list)[0]): <class 'numpy.float64'>
# np.floor(tmp_list): [-4. -3. -2. -1. 0. 1. 2. 3.]


for x in tmp_list:
	print("x:",x,",int:",int(x)) 

# output:
# x: -3.3 ,int: -3
# x: -2.22 ,int: -2
# x: -1.56 ,int: -1
# x: -0.56 ,int: 0
# x: 0.56 ,int: 0
# x: 1.56 ,int: 1
# x: 2.22 ,int: 2
# x: 3.33 ,int: 3

For floating point numbers, int() will truncates toward zero, so -0.1 will by truncate to 0, -1.8 to -1; And its data type will be changed to <class 'int'>.

Averaging histograms

Tue, 29 Dec 2020 00:49:20 +0800

Averaging histograms

An image histogram is the number of each pixel value, which is displayed in the graph.

x axis of the graph is pixel value, range from 0 to 255;

y axis of the graph is the number of this pixel value;

1.How to averaging histograms?

Our goal is to generate a new image with a more even histogram distribution.

How does numpy add two arrays with different shapes?

Sun, 27 Dec 2020 19:54:20 +0800

How does numpy add two arrays with different shapes?

Numpy has a add method which add two numpy array.

Arithmetic operation + does the same thing as Numpy.add;

1.Add a same shapes array

Let’s see a example.

import numpy as np

list1 = np.array([1, 2, 3]);
list2= np.array([10, 20, 30]);

print("list1:",list1,"list2:",list2);
# Print: list1: [1 2 3] list2: [10 20 30]

added_list = list1 + list2;
print("added_list.shape:",added_list.shape,"\nadded_list:",added_list);
# Print:
# added_list.shape: (3,) 
# added_list: [11 22 33]

added_list = np.add(list1, list2);
print("added_list.shape:",added_list.shape,"\nadded_list:",added_list);
# Print:
# added_list.shape: (3,) 
# added_list: [11 22 33]

2.Add a different shape array

But what happen if two array have different shapes?

Understanding Transpose

Tue, 22 Dec 2020 19:55:20 +0800

Understanding Numpy Transpose

1.Transpose is to switch the row and column indices of the matrix A;

x = np.arange(8).reshape((4,2))
print(x)
print(x.T)
# output:
# [[0 1] # x
# [2 3]
# [4 5]
# [6 7]]
# [[0 2 4 6] # x.T
# [1 3 5 7]]

x = np.arange(9).reshape((3,3))
print(x)
print(x.T)
print(x.shape, x.T.shape)
# output:
# [[0 1 2]
# [3 4 5]
# [6 7 8]]
# [[0 3 6]
# [1 4 7]
# [2 5 8]]

x = np.arange(8).reshape((2,4))
print(x)
print(x.T)
# output:
# [[0 1 2 3]
# [4 5 6 7]]
# [[0 4]
# [1 5]
# [2 6]
# [3 7]]

If the array has only one dimension, the transpose of the array will not change;

What do `*args` and `**kwargs` mean in python function?

Tue, 15 Dec 2020 23:21:20 +0800

What do `*args` and `**kwargs` mean in python function?

*args means we can pass an arbitrary number of arguments to the function;

Similarly, **kwargs allow we pass many key=value argument to the function;

`*args`

*args iterable:

def my_sum(*args):
 result = 0
 # Iterating over the Python args tuple
 for x in args:
 result += x
 return result

print(my_sum(1, 2, 3))
# output: 6

The * is a unpacking operator;

def print_three_things(a, b, c):
 print( 'a = {0}, b = {1}, c = {2}'.format(a,b,c))

mylist = ['aardvark', 'baboon', 'cat']
print_three_things(*mylist)
# output: a = aardvark, b = baboon, c = cat

`**kwargs` example

**kwargs iterable:

How Keras add two layers?

Sun, 13 Dec 2020 13:44:00 +0800

How Keras add two layers?

tf.keras.layers.add() method can add two layer?

What it do is sum the values of corresponding positions in two layers.

For example:


input_shape = (1,2,3)
import tensorflow as tf
tf.enable_eager_execution()

print("----------x1 tensor-----------")
x1 = tf.random.uniform(input_shape, maxval=10, dtype=tf.dtypes.int32)
tf.print(x1);
print("----------x2 tensor-----------")
x2 = tf.random.uniform(input_shape, maxval=10, dtype=tf.dtypes.int32)
tf.print(x2);
print("----------add 2 tensors-----------")
y = tf.keras.layers.add([x1,x2])
tf.print(y);

Output: ———-x1 tensor———– [[[7 6 1] [5 7 2]]] ———-x2 tensor———– [[[0 7 8] [2 9 6]]] ———-add 2 tensors———– [[[7 13 9] [7 16 8]]]

Understanding Numpy expand_dims

Mon, 30 Nov 2020 19:55:20 +0800

Understanding Numpy expand_dims

Shape (n,)(n is a number) means it has only one dimension.

The number of values is shape brackets represents the number of dimensions.

import numpy as np
arr = np.array([1,2,3,4,5]);
print("arr shape: ",arr.shape)
print("arr shape: ",arr)
arr2 = np.expand_dims(arr, 0);
print("expand_dims axis=0, shape:",arr2.shape)
print("expand_dims axis=0, arr2:",arr2)
arr2 = np.expand_dims(arr, 1);
print("expand_dims axis=1, shape:",arr2.shape)
print("expand_dims axis=1, arr2:",arr2)

output:

arr shape: (5,) 
arr shape: [1 2 3 4 5]
expand_dims axis=0, shape: (1, 5)
expand_dims axis=0, arr2: [[1 2 3 4 5]]
expand_dims axis=1, shape: (5, 1)
expand_dims axis=1, arr2: [[1]
 [2]
 [3]
 [4]
 [5]]

numpy.expand_dims

expand_dims looks like inserting 1 into the shape brackets base on the axis value;

Sed

Sat, 28 Nov 2020 21:42:20 +0800

Sed

sed is a steam editor. sed treats multiple input files as one long stream.

The full format for invoking sed is:

sed OPTIONS... [SCRIPT] [INPUTFILE...]

Some common OPTIONS:

-n to suppress output, sed -n '45p' file.txt this command prints only line 45 of input file;

-e options are used to specify a script expression, such as sed -e 's/hello/world/' input.txt > output.txt;

-f specify a script file, such as sed -f myscript.sed input.txt > output.txt

Simple Tree command

Sat, 28 Nov 2020 10:15:20 +0800

Simple Tree command

Using the follow command can view current folder tree:

find . -print| sed -e 's;[^/]*/;|____;g;s;____|; |;g'

output

$ find . -print| sed -e 's;[^/]*/;|____;g;s;____|; |;g'
.
|____composer.lock
|____LICENSE
|____README.md
|____.gitignore
|____build-phar.php
|____.git
| |____config
| |____objects
...

Below command only descend at most 2 directory levels:

find . -maxdepth 2 -e -print | sed -e 's;[^/]*/;|____;g;s;____|; |;g'

REFERENCE

Using a Mac Equivalent of Unix “tree” Command to View Folder Trees at Terminal

Categorical Crossentropy源码分析

Wed, 25 Nov 2020 22:48:20 +0800

Categorical Crossentropy源码分析

Source:

import tensorflow as tf
import numpy as np
sess = tf.InteractiveSession()

print("--------output-----------")
target = tf.constant([1., 0., 0., 0., 1., 0., 0., 0., 1.], shape=[3,3])
print("target: \n",target.eval())

output = tf.constant([.9, .05, .05, .05, .89, .06, .05, .01, .94], shape=[3,3])
print("output:\n ",output.eval())

loss = tf.keras.backend.categorical_crossentropy(target, output)
print("loss: \n",loss.eval()) # Output: [0.10536 0.11653 0.06188]

官方文档categorical_crossentropy

Output:

--------output-----------
target: 
 [[1. 0. 0.]
 [0. 1. 0.]
 [0. 0. 1.]]
output:
 [[0.9 0.05 0.05]
 [0.05 0.89 0.06]
 [0.05 0.01 0.94]]
loss: 
 [0.10536055 0.11653383 0.06187541]

问：输出中最后一行，loss第一个值0.10536055是什么得到的？

如何计算RNN和LSTM的参数数量？

Mon, 23 Nov 2020 20:29:20 +0800

如何计算RNN和LSTM的参数数量？

Environment:

python version: 3.7.4
pip version: 19.0.3
numpy version:1.19.4
matplotlib version:3.3.3
tensorflow version:1.14.0
keras version:2.1.5

代码如下:

from keras.layers import SimpleRNN
from keras.models import Model
from keras import Input

inputs = Input((None, 5))
simple_rnn = SimpleRNN(4)

output = simple_rnn(inputs) # The output has shape `[32, 4]`.
model = Model(inputs,output)
model.summary()

Output:

_________________________________________________________________
Layer (type) Output Shape Param # 
=================================================================
input_4 (InputLayer) (None, None, 5) 0 
_________________________________________________________________
simple_rnn_1 (SimpleRNN) (None, 4) 40 
=================================================================
Total params: 40
Trainable params: 40
Non-trainable params: 0
_________________________________________________________________

这里的simple_rnn_1中的param为40是怎么计算的呢？

创建一个简单的RNN网络

Mon, 23 Nov 2020 19:57:20 +0800

创建一个简单的RNN网络

Environment:

python version: 3.7.4
pip version: 19.0.3
numpy version:1.19.4
matplotlib version:3.3.3
tensorflow version:1.14.0
keras version:2.1.5

代码如下：

import keras
from keras import backend as K
from keras.layers import RNN
class MinimalRNNCell(keras.layers.Layer):

	def __init__(self, units,use_bias = True, **kwargs):
		self.units = units
		self.state_size = units
		self.use_bias = use_bias
		super(MinimalRNNCell, self).__init__(**kwargs)

	def build(self, input_shape):
		self.kernel = self.add_weight(shape=(input_shape[-1], self.units), # 添加kernel
										initializer='uniform',
										name='kernel')
		self.recurrent_kernel = self.add_weight(# 添加循环层kernel
			shape=(self.units, self.units),
			initializer='uniform',
			name='recurrent_kernel')

		if self.use_bias:
			self.bias = self.add_weight( # 添加bias
				shape=(self.units,),
				name='bias',
				initializer='uniform',)
		else:
			self.bias = None

		self.built = True

	def call(self, inputs, states):
		prev_output = states[0]
		h = K.dot(inputs, self.kernel)
		output = h + K.dot(prev_output, self.recurrent_kernel)
		return output, [output]


# Let's use this cell in a RNN layer:

cell = MinimalRNNCell(32)
x = keras.Input((None, 5))
layer = RNN(cell)
y = layer(x)

model = keras.Model(x,y)
model.summary()

Output:

Config Github with SSH

Fri, 20 Nov 2020 12:41:20 +0800

Config Github with SSH

Generating a new SSH key

1.Open TerminalTerminalGit Bash.

2.Paste the text below, substituting in your GitHub email address.

$ ssh-keygen -t ed25519 -C "your_email@example.com"

3.Then you’re prompted to do something, press Enter.

Then the keys will be saved under ~/.ssh folder.

You can use ls -al ~/.ssh command to see them.

LOG:

$ ssh-keygen -t ed25519 -C "my_email@example.com"
Generating public/private ed25519 key pair.
Enter file in which to save the key (/c/Users/myusername/.ssh/id_ed25519):
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /c/Users/myusername/.ssh/id_ed25519.
Your public key has been saved in /c/Users/myusername/.ssh/id_ed25519.pub.

Config github key

Copy gitbash:

Quickly host your hugo web on Gitlab

Thu, 19 Nov 2020 13:41:20 +0800

Quickly host your hugo web on Gitlab

Quickly host your hugo web on gitlab.

1.Login gitlab

Gitlab

2.click `new project` after login

Click Create from template in Create new project and use “Hugo template”

In this example, my project name is: testPage

Now you can see your username on Project URL, for example, mine is https://gitlab.com/RhysYao/, and username is RhysYao which will be used later.

telnet

Wed, 18 Nov 2020 19:40:20 +0800

Telnet

telnet host post

示例：

[yaohong@host ~]# telnet www.baidu.com 80
Trying 14.215.177.38...
Connected to www.baidu.com.
Escape character is '^]'.

出现Connected to表示连接上主机；

[yaohong@host ~]# telnet www.baidu.com 882
Trying 14.215.177.38...
telnet: connect to address 14.215.177.38: Connection timed out
Trying 14.215.177.39...

没有出现Connected to表示未连接成功。

如何计算一个BatchNormalization的参数？

Tue, 17 Nov 2020 19:48:00 +0800

如何计算一个BatchNormalization的参数？

# Environment：
# OS			macOS Catalina 10.15.6
# python 		3.7
# pip 			20.1.1
# tensorflow	1.14.0
# Keras 		2.1.5

from keras.models import Sequential
from keras.layers import Conv2D,BatchNormalization
model = Sequential();
# conv2d + max pooling
model.add(
	Conv2D(96, 
		kernel_size = (11,11), 
		strides=(4, 4), 
		padding="valid", 
		input_shape=(224,224,3),
		activation="relu")
	); # output 55 * 55 * 96 

# batchNormalization ! 
model.add(BatchNormalization()) # output
model.summary();

output:

双线插值是什么？

Sun, 15 Nov 2020 22:56:20 +0800

双线插值是什么？

图像处理中，有时我们需要放大图片，比如原来图片宽高是300*300px，如果要在500*500的屏幕上展示，这时一种方法就是把图片直接拉大到500*500，但会发现图像变得模糊了，有没有什么办法可以放大图像而又不会让图像过于模糊呢？

Understand limits to infinity

Thu, 29 Oct 2020 07:57:20 +0800

Understanding limits to infinity

When x->∞, what is the exact number of x?

We don’t know, x is undefined.

∞ is not a number, is a idea of having a greater number than you give.

When x->∞, 1/x appoachs zero, but is never equal zero!

When n -> ∞, does sin(1/n) / (1/n) have meaning?

I used to thought 1/n = 0 while n->∞, but 0 cannot be divided, So I was confused.

SSLCertVerificationError报错

Wed, 28 Oct 2020 23:21:20 +0800

SSLCertVerificationError报错

Error:

ssl.SSLCertVerificationError: [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: unable to get local issuer certificate (_ssl.c:1091)

Solution: 环境：Mac

Macintosh HD > Applications > Python3.6 (或者其它安装python目录）
然后：双击 Install Certificates.command

Double click Install Certificates.command log:

The default interactive shell is now zsh.
To update your account to use zsh, please run `chsh -s /bin/zsh`.
For more details, please visit https://support.apple.com/kb/HT208050.
/Applications/Python\ 3.7/Install\ Certificates.command ; exit;
macdeMacBook-Air-5:~ Rhys$ /Applications/Python\ 3.7/Install\ Certificates.command ; exit;
 -- pip install --upgrade certifi
Requirement already up-to-date: certifi in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (2020.6.20)
WARNING: You are using pip version 20.1.1; however, version 20.2.4 is available.
You should consider upgrading via the '/Library/Frameworks/Python.framework/Versions/3.7/bin/python3.7 -m pip install --upgrade pip' command.
 -- removing any existing file or link
 -- creating symlink to certifi certificate bundle
 -- setting permissions
 -- update complete
logout
Saving session...
...copying shared history...
...saving history...truncating history files...
...completed.
Deleting expired sessions...68 completed.

[Process completed]

Double click Update Shell Profile.command log:

简单的图像加宽和截取类

Mon, 26 Oct 2020 20:15:20 +0800

简单的图像加宽和截取类

源码如下：

class ImageUtils:

	def __init__(self):
		import numpy;
		self.np = numpy;
		pass

	## 
	def resizePadding(self, np_2d_image, target_width,target_height):
		single_img 		= np_2d_image;
		tmp_img_width 	= single_img.shape[1]
		tmp_img_height 	= single_img.shape[0]
		np 				= self.np
		print("resizeFile origin shape :",single_img.shape)
		# 宽度pading添加
		if tmp_img_width < target_width:
			for x in range(tmp_img_width, target_width):
				# tmp_arr = np.arange(255,255,(len(single_img),1));
				if x%2 == 0:
					single_img = np.insert(single_img, 0, 255,axis=1);
				else:
					single_img = np.insert(single_img, single_img.shape[1], 255,axis=1);
				# print(file_name_prefix, "origin shape :",single_img.shape)
		# 高度pading添加
		if tmp_img_height < target_height:
			for x in range(tmp_img_height, target_height):
				# tmp_arr = np.arange(255,255,(len(single_img),1));
				if x%2 == 0:
					single_img = np.insert(single_img, 0, 255,axis=0);
				else:
					single_img = np.insert(single_img, single_img.shape[0], 255,axis=0);

		# 宽度截掉
		if tmp_img_width > target_width:
			for x in range(target_width, tmp_img_width):
				if x%2 == 0:
					single_img = np.delete(single_img, 0, axis=1);
				else:
					single_img = np.delete(single_img, single_img.shape[1]-1, axis=1);

		# 高度截掉
		if tmp_img_height > target_height:
			for x in range(target_height, tmp_img_height):
				if x%2 == 0:
					single_img = np.delete(single_img, 0, axis=0);
				else:
					single_img = np.delete(single_img, single_img.shape[0]-1, axis=0);


		print("resizeFile after shape :",single_img.shape)
		return single_img;

调用代码

如何计算一个卷积层的参数？

Sun, 25 Oct 2020 20:48:00 +0800

如何计算一个卷积参数？

在计算卷积参数前，我们先来看几个小问题。

1.怎么判断一个卷积核有多少通道？

一个卷积核的通道数=前一层的通道数； 比如前一层是64x64x3的图像，那卷积核的通道数就是3层。

2.怎么理解卷积核数量和卷积核通道数？

如果把一个卷积核比喻成一本书，那卷积核的通道数就是一本书的页数。

如何计算全连接神经元参数

Thu, 22 Oct 2020 07:56:20 +0800

如何计算全连接神经元参数？

模型源码为：

from tensorflow.keras import models
from tensorflow.keras import layers
network = models.Sequential()
network.add(layers.Dense(504, activation='relu', input_shape=(504,)))
network.add(layers.Dense(11, activation='softmax'))
network.summary()

输出：

Model: "sequential"
_________________________________________________________________
Layer (type) Output Shape Param # 
=================================================================
dense (Dense) (None, 504) 254520 
_________________________________________________________________
dense_1 (Dense) (None, 11) 5555 
=================================================================
Total params: 260,075
Trainable params: 260,075
Non-trainable params: 0
_________________________________________________________________

在第一行中，dense中的254520是怎么计算呢？其计算方式如下：

Tensorflow 保存和加载model

Wed, 21 Oct 2020 20:56:20 +0800

Tensorflow 保存和加载model

保存model的代码:

import tensorflow as tf
save_model_path 	= "/save_model" # 保存的文件夹路径
network.save(save_model_path);

保存的目录格式如下，期中saved_model.pb是主要的文件：

save_model
	-- assets
	-- variables
		-- variables.data-00000-of-00001
		-- variables.index
	-- saved_model.pb

加载model如下：

一条微信消息在网络层的历程

Wed, 14 Oct 2020 20:40:20 +0800

一条微信消息在网络层的历程

当我们向朋友发送一条微信文字消息时，我们的操作仅仅是让手机连上网，通过连接WIFI或使用移动网络，然后在手机上打字“明天跑步去吗？”，最后点击发送按钮。

PHP autoload

Fri, 14 Aug 2020 20:15:20 +0800

php autoload

In general, when we use a file which not include in current file, we need to load it in current file by using require or include;

But when a project has a lot of php files, it is not a good way to include each file manually.

Let’s see a simple code:

<?php
$person = new Person("Rhys", 20);

var_dump($person);

There is a new class Person which didn’t be load in current file. In this case, php will call default autoload method　spl_autoload() in order to load possible class;

JAVA 序列化

Thu, 28 May 2020 20:26:20 +0800

JAVA 序列化

Java编译过程

Java源码-> javac将源码转换成class 文件->java命令执行 class 文件；

方法中的“ T”用途是什么？

在查看别人的源码时，会看到如下的代码：

Some notes on Convolution Course

Thu, 21 May 2020 15:18:20 +0800

Some notes on Convolution course

What is padding?

Padding is to add some pixels to the border of the original image, such as a 6*6 image will become a 8*8 image if we add a pixel to its border.

valid convolution vs same convolution.

Valid convolution is on padding that means the actual pixels of the output image after we convole original image with filter.

Same convolution means adding padding so that the output image has the same size as its input image.

My Little Baby

Thu, 14 May 2020 22:25:20 +0800

My Little Baby

My wife and I went to hostipal because she felt pain in womb on 6 May. The doctor advised hostipalization for her. She felt pain more frequenctly at that night. Doctors gave her a anesthetic that make her felt less pain.

She was ready to give birth a baby on the next day morning. I was waiting beside the door of delivery room. After two hours, a nurse opened the door and told me that my wife had given birth to a baby successfully. I was so happy at that time.

广东家常菜做法记录

Sun, 03 May 2020 13:43:20 +0800

广东家常菜

豆豉排骨

食材：

排骨一根（约350g）
豆豉（5克约覆盖的碗底）
香菇（10g 约覆盖的碗底）
土豆(80g)
酱油(5ml)
盐(3克)
水（100ml）

器具：电磁炉、电磁炉高压锅

深度学习-第一个数字识别项目

Sun, 03 May 2020 09:15:20 +0800

深度学习-第一个数字识别项目

今天按Google官方推荐流程，整理了开发模板，不是所有深度学习都严格按这个模板来实现，不同项目步骤有所删减，但大体框架是差不多的，主要为以下过程:

深度学习的历史（一）

Fri, 01 May 2020 22:39:20 +0800

深度学习的历史（一）

深度学习的主要突破是从2011年开始，开始超越其它算法成为主流也是从这个时间开始，因此我们有必要先了解从2011年开始的历史，话不多说，直接看。

英语发音-连读和节奏

Mon, 27 Apr 2020 20:38:20 +0800

英语发音-连读和节奏

一、连读

连读指在同个意群里，前一个辅音音素结尾，后一个单词以元音音素开头，把它们连起来读。

连读本质是用更短的时间传达信息，兼顾声音辨别，更加流畅。

Echo set color

Sun, 26 Apr 2020 09:40:20 +0800

Echo set color

先看例子：echo -e "\033[32m Hello world \033[0m";

会输出

$echo -e "\033[32m Hello world \033[0m"
 Hello world # 绿色的文字

其中：

-e表示解析逃逸字符（Escape character），逃逸字符为<esc>作前缀的字符；
\033[32m表示这后文字为绿色；
\033[0m表示清除所有格式设定；

这个语句打印的Hello world文字是绿色的。

How to use CleanWhite Hugo Theme?

Fri, 24 Apr 2020 15:57:20 +0800

How to use CleanWhite Hugo Theme?

CleanWhite Hugo Theme 是Huabing Zhao制作的hugo主题，非常高雅好用，再次感谢Huabing。

我在使用时，花了一点时间，主要遇到以下问题：

1.config.toml怎么配置？
2.怎么在主页显示文章列表？
3.怎么在顶部菜单栏添加分类？
4.主页图片怎么设置？
5.怎么部署到github上？
6.什么设置个人域名？[可选]
7.怎么开启评论？
8.点击右上角的搜索，怎么提示404?

一个一个讲。

Mon, 01 Jan 0001 00:00:00 +0000

The ways to overcome overfitting

Collect data -> Data Processing -> Build model -> Training

 Augmentaion 
 L1,L2 Regularization
 early stopping Regularization
 Dropout
 Reduce Parameters

Mon, 01 Jan 0001 00:00:00 +0000

fit x shape and y shape

# OS:Windows 10
# python version: 3.7.4
# numpy version:1.19.4
# tensorflow version:1.14.0
# keras version:2.1.5

Error

ValueError: Error when checking target: expected dense_2 to have 4 dimensions, but got array with shape (145, 11)

def simpleCNN(input_shape):
	model = Sequential()
	model.add(Conv2D(filters=256, 
							kernel_size = (3*3), 
							activation="relu",
							input_shape=input_shape,
							padding="same" ) );# output: 13*13*384
	model.add(BatchNormalization()) 
	model.add(MaxPooling2D(pool_size=(3, 3), 
								strides=(2,2), 
								padding="same"));

	model.add(Dense(512, activation='relu', ))
	model.add(Dense(11, activation='softmax'))
	# model.add(layers.Dense(11, activation='sigmoid'))
	model.compile(optimizer='rmsprop', loss='categorical_crossentropy',
	 metrics=['accuracy'])
	return model;

model summary:

Mon, 01 Jan 0001 00:00:00 +0000

How to input mathematics formula in markdown?

1.Edit a formula in latex.codecogs.com and download svg suffix file, then store it in you markdown project and refer it with url

![sum_y_substract_ypre](/img/math_formula/sum_y_substract_ypre.svg)

Mon, 01 Jan 0001 00:00:00 +0000

About me

I’m Yaohong.

Try to be a lifelong learner.

Main Url:

https://www.yaohong.vip

https://blog.yaohong.vip （部署在另一台服务器）

Backup url：https://master.d1xvo0d8o10m9n.amplifyapp.com/

Terms in machine learning

Mon, 01 Jan 0001 00:00:00 +0000

Terms in machine learning

FLOPS

FLOPS=Floating point operation per seconds
FLOPs=Floating point operations

REFERNECD: what-is-flops-in-field-of-deep-learning

Yaohong

Comparison Of Programing Language

Comparison Of Programing Language

Constraint, Layout An Size On Flutter, Android, iOS And H5

Layout and Size in Different Systems

Flutter

Android (View System)

iOS (UIKit)

HTML5/CSS/JS

Comparison: Key Aspects

Constraints Handling

Size Determination

Positioning

Flexibility

Examples

Flutter Example:

Android Example:

iOS Example:

HTML/CSS Example:

Summary

What is the purpose of layering the architecture in a Flutter project? How should it be structured?

What is the purpose of layering the architecture in a Flutter project? How should it be structured?

1.The purpose of layering the architecture in a Flutter project

2.What is maintainability?

Must-know information about Swift

Must-know information about Swift

entry point

print log

Variable, Constant

The Flutter Plugin Project Files

The Flutter Plugin Project Files

The Widget tree, Element tree, and Rendering Object tree in flutter

The Widget tree, Element tree, Rendering Object tree in flutter

The project files generated by 'flutter create --template=plugin_ffi' using dart:ffi to call C APIs

The project files generated by ‘flutter create –template=plugin_ffi’ using dart:ffi to call C APIs

The method with the same name from the last mixin will override the previous ones in dart

The method with the same name from the last mixin will override the previous ones in dart

Rhythmically Intermittent When my Iphone 12 Connect To MacOS

Rhythmically Intermittent When my Iphone 12 Connect To MacOS

Xcode or Android Studio is unable to list my iPhone device on MacOS

Xcode or Android Studio is unable to list my iPhone device

Distribution failed with errors When developing ios App

Distribution failed with errors When developing ios App

Flutter HTTP host maven.google.com is not reachable in Windows

Flutter HTTP host https://maven.google.com/ is not reachable in Windows 10

How to handle HTTP host https://maven.google.com/ is not reachable?

1.Get the port of your http proxy

How backward and step are associated with model paramters update?

How are backward and step associated with model paramters update?

nn_Module

nn_Module

1.Where are module parameters configured?

Github ssh key didn't work

Github ssh key didn’t work

Understanding arange, unsqueeze, repeat, stack methods in Pytorch

Understanding arange, unsqueeze, repeat, stack methods in Pytorch

L1 L2 Regularization - Optimizer

Optimizer: L1 L2 Regularization

What does penalize the weights?

How to Label Voice with Praat for Machine Learning

How to Label Voice with Praat for Machine Learning

1.Install

1.1 Download praat

1.2 Install Phonetic symbols

Github API Basic Authentication Example

Github API Basic Authentication Example

1.Generate Personal access tokens

2.Use access token to request Github REST api

Anacode simple usage

Anacode simple usage

1.1 Download and install –Mac os

2.Anaconda Usage

2.1 List all enviroments

2.1 create an enviroment

2.2 activate an enviroment

Simple AI expert Enhanced Loop

Simple AI expert Enhanced Loop

The form of our body in future

The form of our body in future

The Simple Implement of BatchNorm2D

Flutter HTTP host `https://maven.google.com/` is not reachable in Windows 10

How to handle `HTTP host https://maven.google.com/ is not reachable`?