Instagram story viewer> @odey_tech> Posts
Odey Divine(@odey_tech)'s profile avatar HD Profile

Odey Divine

@odey_tech

90
followers
72
following
AI & Data Engineer | Data Science, Machine Learning, GenAI
POSTS STORIES REELS TAGGED
Day 16/50 of my AI Challenge with @victorakinode Machine Learning is built on some fundamentals, which include statistics. Statistics is a branch of mathematics that enables us to understand, analyze, interpret, and present data. Today's lesson was focused on statistics for beginners with Python, and some of the functions we explored are: 1. mean() returns the average of a series 2. std(ddof = 0) returns the standard deviaton of a series 3. median() sorts a series and returns the middle value, etc. Thank you, @victorakinode. It's been an amazing journey so far, learning every day #VAMLChallenge #VAAIChallege #50dayslearningAIandML #victorakinode by @odey_tech
0
18 hours ago
Download
Day 15/50 of my AI Challenge with @victorakinode Turning questions into pictures through data visualisation. Today's lesson focused on data visualisation with Matplotlib. Data visualisation enables us to get a visual summary of our data. There are various charts for different purposes: 1. Line chart for trends over time 2. Bar chart for comparison 3. Histogram for grouping items 4. Scatter plot for visualising pairs, etc. To go a step further, I decided to get a dataset from Kaggle and apply all the concepts being covered so far. I am still working on it Link: https://colab.research.google.com/drive/1VakXXuvNci49Rd3ZObEsILHOJL1Vy11Y?usp=sharing Thank you, @victorakinode. I am not yet where I want to be; I am not where I started, getting 1% better every day #VAMLChallenge #VAAIChallege #50dayslearningAIandML #victorakinode by @odey_tech
0
2 days ago
Download
Day 14/50 of my AI Challenge with @victorakinode Today's lesson focused on Handling Missing Values in Pandas. Data doesn't come in a clean form; it can contain missing values, invalid data types, etc. Handling missing values is also a step in data cleaning; an empty cell doesn't mean the value is zero, and we don't also invent values to avoid our model from inventing predictions. Data imputation is the process of replacing missing values with estimates and Unknown for categorical data. Thank you, @victorakinode. I'm getting 1% better every day #VAMLChallenge #VAAIChallege #50dayslearningAIandML #victorakinode by @odey_tech
0
3 days ago
Download
Day 13/50 of my AI Challenge with @victorakinode Data Cleaning is not a negotiable skill for every data professional; it is very important when working on a data project. Remember, I talked about understanding the problem as part of my review from yesterday's session. After understanding the problem, we then move to collecting the data and understanding it. Today's lesson was focused on Data cleaning with pandas. Data doesn't come in a clean form, although the level of messy data has significantly reduced as a result of organisations recruiting Data Engineers. One rule we must always remember is "garbage-in, garbage-out"; if we train a model with messy data, we are certain to get mediocre results (predictions) Thank you, @victorakinode, for all you do #VAMLChallenge #VAAIChallege #50dayslearningAIandML #victorakinode by @odey_tech
0
4 days ago
Download
Day 12/50 of my AI Challenge with @victorakinode Hands-off Google Colab: What are we building as AI/ML engineers if we don't understand what problem we are trying to solve? Today's lesson was focused on asking good machine learning questions. Business understanding is no longer a negotiable skill for every AI/ML Engineer. Furthermore, not every problem requires machine learning to solve it; that is why we still have Data Analysts and Business Analysts. Even problems that require machine learning don't necessarily require complicated models; sometimes a simple model can be absolutely great in such scenarios. In conclusion, a model tends to predict, but it isn't the same as an explanation of cause. As I continue to advance in my journey, my interest keeps growing in the field of explainable AI Thank you, @victorakinode; this curious builder is getting 1% better every day #VAMLChallenge #VAAIChallege #50dayslearningAIandML #victorakinode by @odey_tech
0
5 days ago
Download
Day 11/50 of my AI Challenge with @victorakinode Actually started feeling like an animal trafficker; why am I importing pandas? 🤣 Today's lesson was not just a continuation of Pandas but was more focused on understanding the dataset before performing any other action. EDA (Exploratory Data Analysis) is the first step every data professional carries out on the dataset Pandas has a lot of methods to explore your dataset, such as .shape to find out the number of rows and columns, .info to provide information about the dataset, and many more Normal routine: a Pandas cheat sheet from DataCamp Link: https://www.datacamp.com/cheat-sheet/pandas-cheat-sheet-for-data-science-in-python Thank you, @victorakinode, for not just putting out this challenge but also for consistently leading by example #VAMLChallenge #VAAIChallege #50dayslearningAIandML #victorakinode by @odey_tech
0
6 days ago
Download
Day 10/50 of my AI Challenge with @victorakinode Today's focus was on working with Tabular data using Pandas. Pandas is a powerful Python library built on NumPy for Data Cleaning and Analysis. Pandas has 2 major components, which are: Series and DataFrame. A Series is more like a column in an Excel spreadsheet, while a DataFrame acts like the entire workbook. Furthermore, we can create our own Series and DataFrame using Pandas, but this becomes really redundant as the dataset grows; instead, we read the already created file through Pandas. For example, we have our customer records saved in a spreadsheet; we can use Python to read the spreadsheet instead of manually typing the spreadsheet content into a DataFrame. Of course, we are not animal traffickers in case you hear "import pandas as pd" Thank you, @victorakinode, for always making time to release gems in the form of lessons #VAMLChallenge #VAAIChallege #50dayslearningAIandML #victorakinode by @odey_tech
0
7 days ago
Download
Day 9/50 of my AI Challenge with @victorakinode Numeric Python (NumPy), the king of numerical computation. Firstly, we can't talk about some Python libraries without tracing their origin back to NumPy. NumPy enables us to carry out computations easily without writing complicated lines of code. I also added a NumPy cheat sheet from DataCamp for anyone interested in learning more about NumPy Link: https://www.datacamp.com/cheat-sheet/numpy-cheat-sheet-data-analysis-in-python Thank you, @victorakinode, for all you do #VAMLChallenge #VAAIChallege #50dayslearningAIandML #victorakinode by @odey_tech
0
8 days ago
Download
Day 8/50 of my AI Challenge Python started making life easy for programmers. Today's lesson focused on functions. Functions are blocks of reusable code. We write the function and call it to enable us to use it. Functions promote one of the principles of clean code, which is DRY (Don't Repeat Yourself). A function is declared with "def" followed by the function name and parameters (optional). Remember from my last post where I mentioned that Python uses indentation instead of the regular curly braces like other languages. Thank you, @victorakinode, for always dropping amazing content, especially on how to debug errors #VAMLChallenge #VAAIChallege #50dayslearningAIandML #victorakinode by @odey_tech
0
9 days ago
Download
Day 7/50 of my AI Challenge Today's lesson focused on making Python smart for decision-making. Imagine you have a security guard at the entrance of your building who makes sure that only authorised individuals are allowed in. We can check if their name is on our record, and this is also possible with Python through conditional statements (if, elif, and else). Secondly, we also looked at loops (for loop and while loop) for repetition of actions instead of repeating the statement several times One of the key takeaways for today's lesson is the fact that Python uses indentation, so we need to be very careful when writing our codes Thank you @victorakinode for always using real-world examples for all the lessons so far Day 7/50 completed, learning never stops #VAMLChallenge #VAAIChallege #50dayslearningAIandML #victorakinode by @odey_tech
0
10 days ago
Download
Day 6 of my AI Challenge Today's lesson focused on Python lists and dictionaries. This is also known as a data structure in programming for effective data storage Python list is a collection of values, and it is denoted by square brackets "[ ]". Imagine we want to store students' test scores for 3 students; instead of having multiple variables for each score, we can declare a list to store them test_score = [30, 17, 19] We can manipulate items contained in the list. Secondly, we have a dictionary, which is a collection of key-value pairs. A significant example is our English dictionary. Whenever we want to search for the meaning of a word, we search for that word, and when we find it, we then retrieve the information it stores student_record = { "name": "Divine", "lesson_completed": 6, "is_active": True } We can also manipulate items in the dictionary This is just Day 6/50, and the journey keeps getting exciting Thank you @victorakinode for all the lessons so far #VAMLChallenge #VAAIChallege #50dayslearningAIandML #victorakinode by @odey_tech
2
11 days ago
Download
Day 5/50 of my AI Challenge Today's lesson was focused on "Python Variables and Data Types". A variable is like a container that gives context to the value it stores. For example, if you see the value 8, what comes to your mind? It might mean the age of a child, maybe the height, or the number of children; we just keep guessing, but with variables we can store the meaning. numbers_of_children = 8 Furthermore, a variable name cannot start with a number; it cannot also contain spaces; instead, we use an underscore. We also need to give our variables meaningful names In addition, Python has 4 major data types, which are: 1. Integer (int): these are whole numbers without decimal points, e.g 17 2. Float (float): these are numbers with a decimal point, e.g 29.8 3. Boolean (bool): this can either be True or False, e.g is_nigerian = True 4. String (str): strings are values that contain quotations; they are used to store text data. Note that when we add quotation marks to numbers, it becomes a string In conclusion, Python is a dynamically typed language. The compiler assigns the data type to the code instead of us statically passing the data type to the variable Thank you so much, @victorakinode, for all the educational sessions so far 10% of the challenge completed, but this journey never ends #VAMLChallenge #VAAIChallege #50dayslearningAIandML #victorakinode by @odey_tech
2
12 days ago
Download

Download All

Save the photos and videos currently loaded on this page.

0Photos
0Videos
Ready to download 0 / 0

Keep this tab open until your download is ready.
back to up