DEV Community: Mohammad Hasibur Rahman

Quantum Properties

Mohammad Hasibur Rahman — Wed, 20 Mar 2024 07:21:43 +0000

Superposition

The way how qubits exist in the state of 0 and 1 at the same time is called Superposition. In definition, Superposition is the ability of a quantum object to be in multiple states at the same time, with probabilities of being measured in each state. All of the states in the superposition are valid and contribute solving a problem

Interference

Interference is how qubits interfere with each other. The ability of waves and wavefunctions to add up or cancel each other out when they overlap is called Interference.

Measurement

The process of forcing a superposition to pick what state the object will be in. This is irreversible and destroys the superposition. The process requires the object to stay in only one of the states. After measuring multiple times, you can also find out at which state the object has the most probability to be in.

Entanglement

Two objects are entangled when one object's state depends on the other object's state.

When two objects are entangled, they have a defined relationship. When you know one object's state, you know something about the other. Entanglement does not mean that two states are the same, it could mean that two states are opposite and it also does not have to be 100% entanglement.

Wavefunction

The wavefunction is how physicists describe what state a quantum object is in, or in other words, everything we can possibly know about it.

For example, here is a visual of the electron's wavefunction at different points in the double slit experiment

[Image taken from Qubit by Qubit course]

Amplitude

The height or size of a wave is called the amplitude. The amplitude of a quantum wavefunction determines how likely it is to be measured.

[Taken from Qubit by Qubit course]

Phase

The phase of a wave is how shifted it is relative to another wave. The phase of a quantum wavefunction plays an important role in interference.

Quantum Computing: Grover's Algorithm

Mohammad Hasibur Rahman — Mon, 19 Feb 2024 06:10:45 +0000

Think if you have a problem set of unordered list and you have many combinations of solutions, how would you solve the problem?

This is where Grover's search Algorithm comes into play. Basically, this algorithm helps you to minimize the combinations of solutions and cut out the ones that are not a part of solution. So you input all the problem set in the Grover's algorithm and it uses a oracle function to process that input and gives you the output. If the output is 1, then you got your solution. If it's 1, then you keep giving your input again and again until you find the result 1.

About "Oracle" function: Amplitude amplification is a general purpose quantum algorithm, or subroutine, that can be used to obtain a quadratic speedup over a handful of classical algorithms. Grover’s algorithm was the first to demonstrate this speedup on unstructured search problems. Formulating a Grover's search problem requires an oracle function that marks one or more computational basis states as the states we are interested in finding, and an amplification circuit that increases the amplitude of marked states, consequently suppressing the remaining states.

Example

The circuit above has a string of 3 bits as an input for a certain problem. The output given depends on whether the input string is a solution to the problem or not.

The result of this checking computation will still be either True or False, but the behaviour of this circuit is slightly different to how you might expect. To use this circuit with Grover's algorithm, we want the oracle to change the phase of the output state by 180° (i.e. multiply by -1) if the state is a solution. This is why Qiskit calls the class 'PhaseOracle'.

Note: PhaseOracle means the Oracle function shifts the phase of the output state by multiplying -1 to the output.

As it shows in the picture, if the output is a solution there is a minus (-) sign in the front of the state. Otherwise, there is not.

Looking at the matrix above, we can see that the rows/columns corresponding to the binary strings 000, 011, and 101 have -1 entries on the diagonal. This means that when the quantum state corresponding to these strings is input into the oracle, the oracle flips the phase of the state (multiplies it by -1), which is a key step in Grover's algorithm for marking the solutions.

The matrix is an 8x8 matrix, which means it can represent all 2^3 (or 8) possible combinations of 3-bit strings. The positions of the -1s in the matrix would be:

000 corresponds to the first position (row 1, column 1),
011 corresponds to the fourth position (row 4, column 4),
101 corresponds to the sixth position (row 6, column 6).

Overview of Grover's algorithm

Now we understand the problem, we finally come to Grover’s algorithm. Grover’s algorithm has three steps:

Here's a quick rundown of my experiment with Grover's Algorithm (Refer to Qiskit documentation if you want to learn more)

Here I initiated the marked states and set the variable "oracle" by calling "grover_oracle" and setting marked_states inside of it.

Next, we can again use Qiskit's tools to create a circuit that does steps 2 & 3 for us.

Now we implement the oracle function in the circuit

Here the marked states means the combinations we are trying on the ciruit to see if it's the solution or not. Here, the marked states are 101,100 & 110. So you can see on the histogram that these states has the highest bars. Quantum computers can have randomness in their results, so it's common to repeat the circuit a few times. This circuit was repeated 1024 times, which is the default number of times to repeat a circuit in Qiskit.

Hope you enjoyed this short blog on grover's algorithm. Leave a comment to support.

Pandas Beginner's Guide

Mohammad Hasibur Rahman — Thu, 11 Jan 2024 10:34:41 +0000

Import the libraries

import numpy as np
import pandas as pd

Series
A Series is very similar to a NumPy array but the only difference is that a Series can have axis labels, meaning it can be indexed by a label, instead of just a number location

Creating a Series

labels = ['a', 'b', 'c']
my_list = [10,20,30]
arr = np.array([10,20,30])
d = {'a':10,'b':20,'c':30}

In the first line of code, I am calling my_list and using the variable data and it prints my_list in a column with indexes from 0 to 2.

In the same way, second line is setting variable data, index for my_list, labels respectively. This specifies the indexes to be the elements of my_list and setting the values to the indexes to labels.

Similarly, you can call the elements from the code cell above without using variables.

Using an Index

The key to using a Series is understanding its index. Pandas makes use of these index names or numbers by allowing for fast look ups of information (works like a hash table or dictionary).

In the first line of code, named the variable "ser1". Then called the Series function where we can add multiple elements. So I added index to the country names and set values to the names 1,2,3,4. In the second line, we called the ser1 variable to print the elements inside the variable.

Same way, we name the variable "ser2" and print the elements inside.

Fifth line: You can call a specific element by calling the name of the element followed by the variable name.

In the last line, you can see I have added both variables ser1 and ser2 to add the eleements together. But when it prints the output, it shows that values for the indexes Italy and USSR has NaN values because they were only elements of either the variable ser1 or ser2

DataFrames
We can think of a DataFrame as a bunch of Series objects put together to share the same index.

Import the randn to load dataset. Name variable df and call the Dataframe function. Inside the function DataFrame I specified the number of
rows(5) and columns(4) and set the indexes to 'A B C D E' and columns to 'W X Y Z'.

Here the split() method is used to split a column into multiple columns in Pandas.

Then we call " df" to print the DataFrame.

Selection and Indexing

You can select each indexes by mentioning the indexes that you want to select followed by the "df" variable. "You may name the variable to something else instead of df. Here df is short for DataFrame."

Creating a new column

Here you can see, a new column has been created where the column name is 'new'. However, the new column is addition of column "W" and "Y", added them together using "+".

Removing Columns

Use df.drop() to drop a column and set the axis to 1. Axis = 1 means you are dropping a column, not a row.

Removing Rows

Similarly, use df.drop() but set the axis = 0 to specify that you are dropping a row. Here, 'E' is the column that you are dropping.

On the second line of code, is another way how you can drop a row. here iloc is short for index location. Specified the row number 2 and dropped it.

On the third line of code is how you can select multiple rows and drop it.

On the fourth line of code, you can select multiple rows and columns. Then you can drop all of them together.

Note: When you are dropping a column, make sure that it won't affect your accuracy rate of the dataset you are working on. There are many ways of handling such columns and rows which has null values.

Conditional Selection

First line: Selected the values greater than 0 which shows the boolean (True or False) values

Second line: Returns the values that are greater than 0 and values lesser than 0 shows NaN.

Third line: Specifies column "W" and drop the values in W that are less than 0. Here you can see, the row "C" has value less than 0. So it dropped row "C".

Microsoft Azure: Cloud concepts and the shared responsibility model

Mohammad Hasibur Rahman — Mon, 13 Nov 2023 01:08:15 +0000

What is cloud computing?

Cloud computing is the delivery of computing services over the internet. Cloud computing services are provided using the internet so it is not depended on physical infrastructure as a physical datacenter.

Shared Responsibility Model

The shared responsibility model shares the responsibilities of maintaining the infrastructure and software needed to keep the datacenter working between its provider and the consumer. Physical security, power, cooling, and network connectivity are the responsibility of the cloud provider and consumer is responsible for the data and information stored in the cloud.
The consumer is also responsible for access security, meaning you only give access to those who need it.

Cloud service types

IAAS (Infrastructure as a service)
PAAS (Platform as a service)
SAAS (Software as a service)

Note: Based on the cloud service types the "Shared responsibility model" varies

The consumer is always responsible for the following:

The information and data stored in the cloud
Devices that are allowed to connect to your cloud (cell phones, computers, and so on)
The accounts and identities of the people, services, and devices within your organization

The provider is always responsible for the following:

The physical datacenter
The physical network
The physical hosts

Depending on the service type responsibility shared:

Operating systems
Network controls
Applications
Identity and infrastructure