Forem: Ochwada Linda

Some Berlin Data on "population density"

Ochwada Linda — Fri, 04 Apr 2025 15:11:49 +0000

Ochwada Linda

Apr 4 '25

Accessing Population Density Data via WFS using Python

#data #wfs #python #gis

Comments

2 min read

Accessing Population Density Data via WFS using Python

Ochwada Linda — Fri, 04 Apr 2025 15:01:22 +0000

For geospatial analysis, demographic context is everything. Whether you're building an AI model for urban planning or visualizing accessibility in city neighborhoods, population data is key. In this post, I’ll walk you through how I downloaded Berlin’s population density data (Einwohnerdichte 2023) using the city’s official Geoportal.

We’ll use Python and OWSLib to connect to a WFS (Web Feature Service) and download the data as a GeoJSON file.

Step-by-Step: Python Function to Fetch WFS Data



from owslib.wfs import WebFeatureService
import geopandas as gpd
import os

def download_demography_data(base_url, typename, output_path):
    os.makedirs(os.path.dirname(output_path), exist_ok=True)

    print(f"🌍 Connecting to WFS: {typename}")

    try:
        wfs = WebFeatureService(url=base_url, version="2.0.0")
        response = wfs.getfeature(typename=typename, outputFormat='application/json')

        gdf = gpd.read_file(response)

        if gdf.empty:
            print("No data found.")
        else:
            gdf.to_file(output_path, driver="GeoJSON")
            print(f"Data saved to: {output_path}")

    except Exception as e:
        print(f"Failed to download: {e}")

What’s Happening Here?

WebFeatureService: Establishes a connection to the WFS endpoint.

getfeature: Retrieves the specified layer in GeoJSON format.

GeoPandas: Loads the response and saves it locally.

os.makedirs: Ensures the output directory exists.

WFS URL and Available Layers


if __name__ == '__main__':

url = "https://gdi.berlin.de/services/wfs/ua_einwohnerdichte_2023"
wfs = WebFeatureService(url=url, version="2.0.0")

print("\nAvailable Layers:")
for layer in list(wfs.contents):
    print("🔹", layer)

You’ll see something like:

🔹 ua_einwohnerdichte_2023:einwohnerdichte2023

Downloading the Layer


# In main

# Downloading Berlin's 2023 population density layer
base_url = "https://gdi.berlin.de/services/wfs/ua_einwohnerdichte_2023"
layer = "ua_einwohnerdichte_2023:einwohnerdichte2023"
output_path = "data/berlin_population_density.geojson"

download_demography_data(base_url, layer, output_path)

Why This Matters

This data is incredibly useful for:

Urban planning
Accessibility mapping
Infrastructure development
AI/ML models that incorporate population variables

You now have a clean GeoJSON file with up-to-date demographic data, ready for QGIS, Python analysis, or integration into a web map.

👉 More on Blogs
👉 Reach Out

Fetching and Loading Data from Github

Ochwada Linda — Thu, 29 Dec 2022 17:58:25 +0000

It's usually preferable to write a function that downloads and decompresses data from github (or online data) rather than doing it manually.

This is especially true if the data changes frequently: you can write a small script that uses the function to retrieve the most recent data (of you can set up a scheduled job to do that automatically at regular intervals). Automating the data retrieval process is also useful if you need to install the dataset on multiple machines.

Here is the fuction to fetch and load data:

# ----- Libraries ---------
from pathlib import Path

import pandas as pd
import tarfile
import urllib.request
# --------------------------

# Function to fetch and load data --->
def function_name():
    zipped_path = Path("datasets/files.tgz")

    if not zipped_path.is_file():
        Path("datasets").mkdir(parents=True, exist_ok=True)
        url = "https://github.com/****/Datasets/raw/main/files.tgz"
        urllib.request.urlretrieve(url, zipped_path)

        with tarfile.open(zipped_path) as file_name:
            file_name.extractall(path = "datasets")

    return pd.read_csv(Path("datasets/files/file.csv"))

data_file = function_name()

When function_name()is called, it will look for the dataset file in datasets/files.tgz. If it does not find it, it will create a directory datasets inside your working directory; then it will download the files.tgz from the site https://github.com/****/Datasets/raw/main/files.tgz. This files.tgz contains the file file.csv.

The function with then lod the CSV file into a Pandas DataFrame object containing all the data, and return it.

You can check your data by:

print(data_file[:10])
 # Or 
print(housing.head())

To display the top 10 rows ( or 5 top rows) of your data.

A look into Machine Learning in Sales - What do Dealcode’s models do?

Ochwada Linda — Tue, 30 Aug 2022 12:07:15 +0000

According to a Harvard Business Review study, data-driven decision-making improves business performance. Data-driven decision-making refers to the practice of making decisions based on data analysis rather than intuition. These data-driven companies that pioneered the use of AI in sales are overjoyed with the results, which include a more than 50% increase in leads and appointments, 40%-60% cost savings, and 60%-70% call time savings. When the benefit of allowing sales professionals to spend more time completing transactions is considered, AI's allure becomes even more compelling.

Machine Learning and Artificial Intelligence, a general view

AI and machine learning have the potential to eliminate the most time-consuming, manual activities that prevent sales teams from spending more time with clients — automating marketing processes, using predictive analytics, forecasting, reporting, and advising, all of which help clients upsell. All of these are methods for reducing labor in sales teams. When using data-driven decision-making, businesses in the top third of their industry are 5% more productive and 6% more profitable than their competitors (McAfee and Brynjolfsson, 2012).

Caruana and Niculescu-Mizil (2006) examined the top machine learning models with higher performance as well as those that aren’t as significant and are commonly used in sales. The best machine learning (ML) models, such as random forests, boosting, support vector machines (SVM), and neural networks, are demonstrated in this study. In terms of predictive performance, these models outperform simple, interpretable models such as decision trees, naive Bayes, and decision rules. Companies that do not use the aforementioned open-sourced models use their own patent AI and ML models.

Dealcode - the AI Selling Software.

Dealcode is an AI Guided Selling Software that extracts data from CRMs and processes it using its own AI and machine learning model. It is a predictive analytics tool that predicts the likelihood of success for sales pipeline prospects and risks. It gives sales teams the most up-to-date information on which deals to prioritize and who to contact right away. This predictive analytics is performed by analyzing sales processes using a patented machine learning model. Dealcode identifies individual factors that contribute to the success or failure of a sales team. Sales are noticeably more effective as a result. It also saves money on expensive resources for complex data analyses.

What services does Dealcode offer? Read more on Dealcode’s capabilities from my blog and company’s website - link.

Deal Scoring- Part 1: Data Pre-processing in MySQL

Ochwada Linda — Thu, 23 Jun 2022 11:34:24 +0000

In my last blog, I wrote about AI & Sales; today, I will start the process of sales data analytics. I will be building a logistic regression model to assign a score to each of the prospects that a company can use to target potential deals. A higher score would mean that the deal is hot, i.e. is most likely to convert, whereas a lower score would mean that the deal is cold and will mostly not get converted.

When closing deals, a salesperson with a rich pipeline of potential clients must decide where to focus their time to close a particular deal. Often, salespeople make decisions based on their intuition and incomplete information. Using AI, data scientists compile historical information about a client, company, social media postings and the salesperson’s customer interaction history (e.g. emails sent, calls made, text sent etc. ); and rank the opportunities or leads in the pipeline according to their chances (probability) of closing successfully. One tool built on this methodology is Dealcode GmbH.

The data obtained was in several CSV files; I used MySQL to inspect and preprocess the data.

-- Creating NEW Table
DROP TABLE IF EXISTS crm_activity;

CREATE TABLE crm_activity (
    id INT PRIMARY KEY,
    crmId INT,
    crmUserId INT,
    crmDealId INT,
    crmLeadId INT,
    crmDuration TIME NULL,
    crmType VARCHAR(255),
    crmDueTime TIME NULL,
    crmAddTime VARCHAR(255),
    crmUpdateTime VARCHAR(255),
    crmDueDate DATE,
    crmSubject VARCHAR(255),
    crmDone INT,
    isDeleted INT,
    createDate VARCHAR(255),
    companyId INT
);

LOAD DATA LOCAL INFILE 'path/crm_activity.csv' IGNORE 
INTO TABLE crm_activity
FIELDS TERMINATED BY ',' 
LINES TERMINATED BY '\n';

SELECT * FROM crm_activity
LIMIT 100;

Data type at a field like createDate is presented as character data that is varying (VARCHAR). Convert the data type to a timestamp using str_to_date, and insert the converted string-to-date in a new column. The string date-time column can then be deleted by using DROP.

ALTER TABLE crm_activity ADD (
    crmAddTime_ts TIMESTAMP,
    crmUpdateTime_ts TIMESTAMP,
    createDate_ts TIMESTAMP
);

UPDATE crm_activity SET  crmAddTime_ts = str_to_date( crmAddTime, '"%Y-%m-%d %H:%i:%s"');
UPDATE crm_activity SET crmUpdateTime_ts = str_to_date( crmUpdateTime, '"%Y-%m-%d %H:%i:%s"');
UPDATE crm_activity SET  createDate_ts = str_to_date(  createDate, '"%Y-%m-%d %H:%i:%s"');

DELETE FROM crm_activity WHERE id=0;

COMMIT;

MySQL automatically generates a database schema view from the created (now existing) database (From the interface --> Database --> Reverse Engineer --> Follow the instructions).

Database Schema Generated

Database schemas are essential because they help developers visualize how a database should be structured and give a clear point of reference about what tables and fields a project contains.

Data Processing

Fetching data from MySQL into Anaconda (python), requires mysql.connector package, installed it by pip install mysql-connector-python

mydb = mysql.connector.connect(
  host="localhost",
  user="myuser",
  password="yourpassword",
  database="name_of_the_database"
)

# Loading and visualizing in Anaconda
crm_deal = mydb.cursor()
crm_deal.execute("SELECT * FROM crm_deal ")
crm_deal_results = crm_deal.fetchall()
for leads in crm_deal_results:
  print(leads[:10])

Continue reading - Deal Scoring - Part 2: Data processing with Python

AI for Sales

Ochwada Linda — Fri, 27 May 2022 10:16:38 +0000

Smart technology has revolutionised the world as we know it and gets smarter with each passing hour; where do salespersons stand? How is technology changing the way salespeople do business? What are the unseen consequences in sales or business if salespersons do not adapt to the Fourth Industrial Revolution? Like many other industries, the world of sales and marketing is experiencing a significant change - ‘tectonic shifts. Thanks to the cutting edge digital infrastructure, Artificial Intelligence (one of the advances of the Fourth Industrial Revolution) has a global footprint, I.e. in every industry. According to a study conducted by Harvard Business Review, companies using AI in sales increased their leads by more than 50%, reduced call time by 60-70 %, and realised cost reductions of 40-60%.

Artificial Intelligence has revolutionised sales departments in companies, and at present, companies boost their lead volume, improve rate, and supercharge overall sales performance. AI algorithms can help power any company’s decision-making process and make better business predictions.

According to “Sales Ex Machina - How Artificial Intelligence is Changing the World of Selling” by V Antonio and J. G Anderson, PhD, there are five specific areas where AI algorithms can be leveraged by sales teams hence growing your businesses. - Lead Scoring, Price Optimisation, forecasting, upselling and cross-selling, and managing performance (Using this book as a basis for my research)

Deal Closure and Lead Scoring

When closing deals, a salesperson with a rich pipeline of potential clients have to decide where to focus their time to close a particular deal. Often, salespeople make decisions based on their intuition and incomplete information. Using AI, data scientists compile historical information about a client, company, social media postings and the salesperson’s customer interaction history (e.g. emails sent, calls made, text sent etc. ); and rank the opportunities or leads in the pipeline according to their chances (probability) of closing successfully. One tool built on this methodology is Dealcode GmbH with the slogan TIME KILLS DEALS!

Dealcode is an AI guided selling software tool that determines the winning probability of the prospectus as well as the risk involved in a sales pipeline. Using data from Hubspot and Pipedrive and information from a web crawler, it provides sales team with up-to-date information on what deals they should focus on and whom they should talk to ‘today’. Their patented machine learning model analyses over 150 attributes to determine individual factors that contribute to the success or failure of the sales team, hence making the sales measurably more effective

A web crawler is an internet bot that systematically browses the world wide web using keywords.

Price Optimisation

Using AI to tell what an ideal discount rate should be for a proposal to ensure that you’re most likely to win the deal by looking at specific features of each past deal won or lost. These features may include the size of the deal, product specifications, number of competitors, company size etc.

Demand Forecasting

Using AI to forecast clients’ demands by analysing past data. This helps companies optimise their supply chain, sales and marketing operations, preventing under-stocking or overstocking.

Upselling and cross-selling

Using AI to identify which of your existing clients would buy a better version of what they currently own and/ or which- would likely want to buy a new product offering altogether, increasing revenue and a drop in the marketing costs.

Managing Performance

Using AI to identify the likelihood of a salesperson hitting their quotas and which outstanding deal stands a good chance of being closed. This allows managers to focus their attention on key salespeople and associated deals that would help the company hit their quota.

The accuracy of the predictions depends on the system and the quality of the data used; the AI systems need the proper inputs from the past and the present to draw conclusions (insights) and create predictions.

Companies always look for ways to increase their revenue, cut back on costs, and expand market share while significantly minimising risks. Leveraging existing information on clients, vendors, and web crawling enable leading companies to analyse and generate actionable insights that help their sales team close more deals.

SQL Basic - What you need to know! (Basics)

Ochwada Linda — Wed, 11 May 2022 06:18:16 +0000

Main components of SQL’s Syntax:

Data Definition Language (DDL)

A set of statements that allow user to modify data structures and objects e.g. tables

The CREATE statement

CREATE object_type object_name;

CREATE TABLE object_name(column_name data_type);

CREATE TABLE sales(purchase_no INT);

The ** ALTER** statement Used when altering existing objects; ADD, REMOVE, RENAME

-- Add a column in a table
ALTER TABLE sales
ADD COLUMN date_of_purchase DATA;

-- delete an entire table   
DROP TABLE customer;

-- Rename a table   
RENAME TABLE sales TO customer;

The TRUNCATE statement Instead of deleting an entire table through DROP, you can remove its data and continue to have as an abject in the database.

Data Manipulation Language (DML)

Statements that allow to manipulate the data in the tables of a database.

The SELECT statement SELECT … FROM..

-- deliver all the records in the table.
    SELECT* FROM sales; 
-- Select specific specific data from the database. 
SELECT row FROM sales;

The INSERT statement
Used insert data into tables INSERT .. INTO .. VALUES …;

-- Choose specific cell to insert values
INSERT INTO sales (purchase_number, date_of_purchase) 
VALUES(1, “12-09-2020”);

-- Insert value
INSERT INTO sales VALUES(1, “12-09-2020”);

The UPDATE statement UPDATE.. SET.. WHERE Allows you to renew existing data of your table

UPDATE sales
SET date_of_purchase = “16-11_2020”
WHERE purchase_number =1;

The DELETE statement DELETE… FROM… WHERE Specify what is to be removed in a table.

DELETE FROM sales
WHERE purchase_number =1;

Data Control Language (DCL)

GRANT and REVOKE statements. They allow rights of people using the database and used by administrators.

The GRANT statement Gives or grants certain permissions to users.

GRANT type_of_permission ON database_name.table_name TO ‘username’@‘localhost’

-- Creates user `ochwada` whose password is "password"
CREATE USER ‘ochwada’@‘localhost’ IDENTIFIED BY ‘password’;

-- Allow the user to use all SELECT Statements on the table 'customer' from the 'sales' database.
GRANT SELECT ON sales.customers To ‘ochwada’@‘localhost’;

-- Use all statements and all tables in the 'sales' database.
GRANT ALL ON sales.* To ‘ochwada’@‘localhost’;

The REVOKE Clause Used to revoke permissions and privileges of database users.

REVOKE type_of_permission ON database_name.table_name TO ‘username’@‘localhost’

-- Only the use of SELECT statement will be REVOKED for the table customers in the sales database.  
REVOKE SELECT ON sales.customers To ‘ochwada’@‘localhost’;

-- use of all statements in the 'sales' database will be revoked.   
REVOKE ALL ON sales.* To ‘ochwada’@‘localhost’;

Transaction Control Language (TCL)

Not every change you make to a database is saved automatically.

The COMMIT statement Related to INSERT, DELETE, UPDATE. It will save the changes you’ve made permanently Allow other users to have access to modified version of the database.

UPDATE sales
SET date_of_purchase = “16-11_2020”
WHERE purchase_number =1;

COMMIT;

The ROLLBACK Clause The clause that will let you make a step back . Allow you to undo any changes you have made but don’t want to be saved permanently .

UPDATE sales
SET date_of_purchase = “16-11_2020”
WHERE purchase_number =1;
COMMIT;

ROLLBACK;

Get Started with Face Detection with Python

Ochwada Linda — Mon, 02 May 2022 19:35:57 +0000

NOTE: Used a pre-trained model
Face detection in photos can be performed using the classical feature-based cascade classifier using the OpenCV library (cv2).

Used packages

import cv2
import time
import os

OpenCV provides the CascadeClassifier class that creates a cascade classifier for face detection. The constructor can take a filename as an argument that specifies the XML file for a pre-trained model.

Download a pre-trained model (file here) for frontal face detection from the OpenCV GitHub project - to be placed in your working directory.

Setup the image directory

# function to get images from folder 
def get_images(dir_name):
    list_images = os.listdir(dir_name)
    all_images =list()

    for entry in list_images:
        full_path =os.path.join(dir_name, entry)
        if os.path.isdir(full_path):
            all_images.all_images + get_images(full_path)
        else: all_images.append(full_path)
    return all_images

Load the model that will perform face detection on photographs by calling the detectMultiScale() function.

# Face Detection

def  main():
    dir_name = 'images' # directory for images
    list_images = get_images(dir_name)

    for i in range(20): #20 images
        image_path = list_images[i]

        print(image_path)

        # load the pre-trained model
        case_path = "haarcascade_frontalface_default.xml" # define the model used for recognition detection
        faceCascade= cv2.CascadeClassifier(case_path)
        image=cv2.imread(image_path)
        gray = cv2.cvtColor(image,cv2.COLOR_BGR2GRAY) # make the picture to gray from color

        # face detection 
        faces = faceCascade.detectMultiScale(gray,
                                            scaleFactor = 1.1,
                                            minNeighbors=5,
                                            minSize = (30, 30))

        for (x, y, w, h) in faces: # draw rectangles on the faces when detected
            cv2.rectangle(image, (x,y), (x + w, y+h), (0, 255, 0), 2)

        #Load the detected faces
        cv2.imshow("Face Found", image)
        cv2.waitKey(5)
        time.sleep(5)
        cv2.destroyAllWindows()

if __name__ == '__main__':
    main()

Running the model loads the images and configures the cascade classifier; faces are detected, and each bounding box gets printed.

The model only works for faces directly at the camera (in front).

2 Python Project Ideas for Beginners

Ochwada Linda — Sun, 01 May 2022 09:43:11 +0000

1. Text Detection and Extraction using OpenCV and OCR.

OpenCV (Open source computer vision) is a library mainly aimed at real-time computer vision, in python it helps to process an image and apply various functions e.g. pixel manipulations, object detection, etc. For this project using contours to detect the text in an image and save it to a text file.

Requirements

conda install -c conda-forge opencv
conda install -c conda-forge pytesseract

Coding

I will use the following sample image:

import cv2
import numpy as np
import pytesseract as pyt

image = cv2.imread('The_Zen_of_Python.jpeg')

# image convertion to grey scale
image_gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
image_gray, binary_image = cv2.threshold(image_gray, 128,255, cv2.THRESH_BINARY)
image_gray =cv2.bitwise_not(binary_image)

kernel = np.ones((2,1), np.uint8)
image = cv2.erode(image_gray, kernel,iterations = 1)
image = cv2.dilate(image_gray, kernel,iterations = 1)

The next code will convert the image into a string and print the string as text.

# image conversion into string
image_text = pyt.image_to_string(image)

print(image_text)

Check out the video of the project above -Project video

2. Stock Price Visualisation using Yahoo's yfinance .

yfinance is an open source library developed by Ran Aroussi as a means to access the financial data available on Yahoo Finance.

Yahoo Finance offers a wide range of market data on stocks, bonds, currencies and cryptocurrencies. In addition it offers market news, reports and analysis and fundamentals data- setting it apart from some of it’s competitors.

Requirements

pip install yfinance
conda install -c conda-forge pandas
pip install plotly-express

Coding

import yfinance as yf
import pandas as pd
import datetime 
from datetime import date, timedelta

today = date.today()

date1 = today.strftime("%Y-%m-%d")
end_date = date1

date2 = date.today() -timedelta(days = 365)
date2 = date2.strftime("%Y-%m-%d")
start_date = date2 

data = yf.download ( 'AAPL', 
    start = start_date, 
    end = end_date,
    progress = False)
print(data[:10])

Visualise your Stock price data using Plotly Express. This visualisation will prompt a new window to open in your browser with up todate data.

import plotly.express as px 
fig = px.line(data, x = data.index, 
                y = "Close", 
                title = "Stock Price data")
fig.show()

SQL Basic Cheat Sheet

Ochwada Linda — Sat, 30 Apr 2022 20:06:24 +0000

Structured Query Language (SQL), is a programming language that is used to manage relational databases, and perform operations on the data in them.

Sample Data

1. SIMPLE QUERYING

-- Fetch / retrieve all columns from a table.
SELECT *
FROM table_name;

-- Fetch columns from the table and order the results.
SELECT col1, col2
FROM table_name
ORDER BY col1 DESC;

ALIASES

-- COLUMNS
SELECT col1 AS col_11
FROM table_name;

-- TABLES
SELECT col1, col1_1, col1_2
FROM table_name AS t1
JOIN name_table AS t2
ON t1.col1 = t2.col1_1;

2. FILTERING THE OUTPUT

Comparison

SELECT *
FROM table_name
WHERE col1 BETWEEN range1 AND range2; 

-- aggregate functions
SELECT *
FROM table_name
WHERE col1 > 3;

TEXT OPERATORS

-- List with Names starting with Mar
SELECT * 
FROM employees
WHERE first_name LIKE ('Mar%');

-- list with NAMES ENDING WITH 'rk'
SELECT * 
FROM employees 
WHERE first_name LIKE ('%rk');

-- Others
SELECT  * 
FROM employees
WHERE first_name
NOT IN ('John' , 'Mark', 'Jacob');

3. QUERYING MULTIPLE TABLES

-- CROSS JOIN 
-- Connecting two tables
SELECT
    dm.*, d.*
FROM
    dept_manager dm
     CROSS JOIN
    departments d
ORDER BY dm.emp_no, d.dept_no;  

-- INNER JOIN 
-- Null values are also not displayed
SELECT m.dept_no, m.emp_no, d.dept_name
FROM dept_manager_dup m
INNER JOIN
departments_dup d on m.dept_no = d.dept_no
GROUP BY m.emp_no
ORDER BY m.dept_no;

-- LEFT JOIN (interchangeable with LEFT OUTER JOIN)
-- All matching values of the two tables + all values from the left table that matches no values
SELECT m.dept_no, m.emp_no, d.dept_name
FROM dept_manager_dup m
    LEFT JOIN departments_dup d ON m.dept_no = d.dept_no
GROUP BY m.emp_no
ORDER BY m.dept_no;


-- RIGHT JOIN
-- Identical to Left JOIN, with the only difference being that the direction of the operation is inverted. 
SELECT m.dept_no, m.emp_no, m.dept_name
FROM dept_manager_dup m,
     departments_dup d
WHERE m.dept_no = d.dept_no
ORDER BY m.dept_no DESC;

-- SQL Self Join
-- Applied when a table must join itself (Combining rows OF A TABLE with other rows of same table)
-- From the emp_manager table, extract the record data only of those employees who are managers as well.
SELECT *
FROM emp_manager em
ORDER BY em.emp_no;
SELECT em1.*
FROM emp_manager em1
    JOIN emp_manager em2 ON em1.emp_no = em2.manager_no;

4. SUBQUERIES

SQL Sub-queries with IN Nested Inside WHERE Queries inside queries
also called inner queries/nested queries/inner select/ outer select.

SELECT * FROM dept_manager;

SELECT e.first_name, e.last_name
FROM employees e
WHERE e.emp_no IN (
        SELECT dmd.emp_no
        FROM dept_manager_dup dmd
    );

5. SET OPERATIONS

Set operations are used to combine the results of two or more queries into a single result. The combined queries must return the same number of columns and compatible data types. The names of the corresponding columns can be different.

-- UNION
 SELECT
    e.emp_no,  e.first_name, e.last_name,
    NULL AS dept_no, NULL AS from_date
FROM employees_dup e
WHERE e.emp_no = 10001
UNION  -- difference
    SELECT 
        NULL AS emp_no, NULL AS first_name,
        NULL AS last_name, dmd.dept_no,
        dmd.from_date
FROM dept_manager_dup dmd;

-- UNION ALL
 SELECT
    e.emp_no,  e.first_name, e.last_name,
    NULL AS dept_no, NULL AS from_date
FROM employees_dup e
WHERE e.emp_no = 10001
UNION ALL -- difference
    SELECT 
        NULL AS emp_no, NULL AS first_name,
        NULL AS last_name, dmd.dept_no,
        dmd.from_date
FROM dept_manager_dup dmd;

NOTE: I highly recommend MySQL for Data Analytics and Business Intelligence tutorial for SQL beginners with focus on Data analysis.

How to Switch to the default GPU for High Performance Graphic Card in Windows 11

Ochwada Linda — Sun, 24 Apr 2022 23:31:14 +0000

When using computers, we tend to utilise the in-built integrated graphics processors that help with providing basic features at a low power consumption.

Sometimes, you may need to use apps (or games) that require high processing power. Windows 11 automatically switches between graphics cards, depending on the apps you are using and the games you’re playing. Though at times you may need (or be forced) to switch between a discrete video card (dedicated for high processing power) and your inbuilt processor.

Why the need to Switch between default GPU for High Performance GPU?

When working on normal apps e.g. Excel, the integrated graphic card chip is powerful enough to handle the program. However when doing projects in Data Science and also "heavy gaming" ( for example: RIDE 4)

Recipe for a computer meltdown without a High Performance GPU!

Normally there is an automatic switch between the two graphics cards; but with apps or programs with bugs, you will be forced to change the GPUs manually.

Below are steps to follow for SWITCHING the graphic cards:

Step 1. Open the Settings app. On System window --> click on Display.

Step 2. From the Display window, scroll down to Graphics.

Step 3. On the Graphics page, you can either choose between adding an app from Desktop app or Microsoft app or simply Search for the app that needs the High Performance GPU.

Step 4. Click on the app, --> Options button.

Step 5. A dialog box will pop up (Graphics preference) Click on the preferred graphical Card and then save.

Traffic Light Simulator - Arduino

Ochwada Linda — Sat, 23 Apr 2022 08:57:09 +0000

Simulating a traffic light using LEDs and Arduino, I did the project with my 8-year-old son! He found it so cool.

COMPONENTS

Arduino UNO
Breadboard (Generic)
Jumper Wires (Generic)
LED (Red, Green, Yellow)
Arduino USB 2.0 Data Cable
Resistor 100 ohm

COMPONETS SETUP

IDE used: Arduino IDE

This simple "for beginners " uses an Arduino and LEDs to simulate a traffic light.

Define variables, use lights by colour name:

// variables
int red = 10;
int yellow = 9;
int green = 8;

Create a function to configure the output colours of the LEDs (Red, Yellow, Green):

void setup(){
    pinMode(red, OUTPUT);
    pinMode(yellow, OUTPUT);
    pinMode(green, OUTPUT);
}

This pinMODE function, configures the Arduino to use a defined pin as an output.

Write the code for the actual logic behind the traffic light simulation. The given values are in milliseconds

void loop(){
    changeLights();
    delay(15000);
}
void changeLights(){
    // green off, yellow on for 3 seconds
    digitalWrite(green, LOW);
    digitalWrite(yellow, HIGH);
    delay(3000);

    // turn off yellow, then turn red on for 5 seconds
    digitalWrite(yellow, LOW);
    digitalWrite(red, HIGH);
    delay(5000);

    // red and yellow on for 2 seconds (red is already on though)
    digitalWrite(yellow, HIGH);
    delay(2000);

    // turn off red and yellow, then turn on green
    digitalWrite(yellow, LOW);
    digitalWrite(red, LOW);
    digitalWrite(green, HIGH);
    delay(3000);
}

Upload the above code to your Arduino and run (make sure the right board and port are selected from Tools menu)