MIB | Machine Intelligence Blog

383 FOLLOWERS

Up & running with machine learning | semasuka stern machine learning blog

MIB | Machine Intelligence Blog

1y ago

Hello Folks! Welcome back. In this post, I will discuss the theory behind linear regression models, one of the wildly used machine learning models to predict continuous variables (fancy terms to say that we are predicting a number, also referred to as numerical target or label). The model is quite simple to understand yet powerful. We use it when model interpretability (when we want to know which dependent variables, aka features, are the most predictive) is required, like in the consumer lending or medical fields where transparency is at its core. In the next post, we will dive deeper into th ..read more

MIB | Machine Intelligence Blog

2y ago

Welcome back, forks! After a long period of not posting here, I am happy to share that I am back again on MIB. In this post, we will work on an end-to-end machine learning project. I firmly believe this is one of the most detailed and comprehensive end-to-end ML project blog post on the internet. This project is perfect for the beginner in Machine Learning and seasoned ML engineers who could still learn one or two things from this post. This project was featured on Luke Barousse Youtube channel, click here to watch the video.
Here is the roadmap we will follow:
We will start with exploratory ..read more

MIB | Machine Intelligence Blog

3y ago

Welcome back! I am very excited about this post as we are introducing machine learning and its commonly used jargon. You will have a broad overview of machine learning, how it works, and even write our first machine learning code at the end of the post. To understand advanced machine learning, we first need to have a good grasp of the fundamentals. That is why I think this is the most important post on this blog so far.
With no further due, let’s get started.
1. What is machine learning, and how can a machine learn? 1.1 What is machine learning
Machine learning is a subfield of computer scienc ..read more

MIB | Machine Intelligence Blog

3y ago

Great to see you again here! In this last post of the Pandas series, we will continue exploring advanced DataFrame exercises. Pandas is easer to learn than NumPy, in my opinion. Its documentation is well written, so don’t be shy! Read its documentation throughout if you get stuck here.
Let’s get started by importing NumPy and Pandas.
import numpy as np
import pandas as pd
Ex 51: How to get the row number of the nth largest value in a column?
Q: Find the row position of the 5th largest value of column a of the_dataframe.
from numpy.random import default_rng
np.random.seed(42)
rng = defaul ..read more

MIB | Machine Intelligence Blog

4y ago

Welcome back, guys! We will continue with part 2 in this series of Pandas exercise. I am very excited about this post because we will introducing DataFrame, the most used Pandas data structure. I hope you guys will enjoy this post.
With no further due, let’s get started.
We will start by importing Pandas and NumPy
import pandas as pd import numpy as np
Ex 26: How to get the mean of a series grouped by another series?
Q: Compute the mean of weights of each fruit.
fruits = pd.Series(np.random.choice(['apple', 'banana', 'carrot'], 10)) weights = pd.Series(np.linspace(1, 10, 10))
Desired o ..read more

MIB | Machine Intelligence Blog

4y ago

Welcome back, folks! In this series of 3 blog post, we will be discussing pandas which one of my favorite python libraries. We will go through 74 exercises to solidify your skills with pandas and as usual, I will explain the WHY behind every single exercise.
Pandas is a powerful open-source library for data analysis and data manipulation. The library is packed with a ton of feature, well supported and documented by the community. It is built on top of NumPy and integrate well with all the main machine learning libraries like Scikit-learn and Matplotlib.
Pandas already come bundles in the Anaco ..read more

MIB | Machine Intelligence Blog

4y ago

Welcome back, folks! This post is the last in the series of NumPy exercises. In this post, we will see intermediate and advanced level exercises. Remember, the more you practice, the more you will understand NumPy and will use it with ease in your ML projects.
Let’s now start by importing the NumPy as usual.
import numpy as np
Ex 41: Convert a numeric to a categorical (text) array
Q: Bin the petal length (3rd) column of the_data_2d to form a text array, such that if petal length is:
Less than 3 we change it to ‘small.’
between 3 and 5 we change it to ‘medium.’
Greater or equal to 5 we cha ..read more

MIB | Machine Intelligence Blog

5y ago

For this second post of NumPy exercises series, we will be doing intermediate level exercises in NumPy and will go through the solution together as we did in the first part. Try to solve the exercises on your own then compare your answer with mine. Let’s get started.
We first import the NumPy.
import numpy as np
Ex 21: Create a two-dimensional array containing random floats between 5 and 10
Q: Let’s create a two-dimensional array with a shape of 5x3 that contain random decimal between 5 and 10.
Desire output
# Of course the numbers in the array will not be the same as mine, but your sol ..read more

MIB | Machine Intelligence Blog

5y ago

In this post, we will be solving 20 exercises in NumPy to sharpen what you have learnt from the NumPy introduction post. If you have not read the NumPy post, I highly encourage to go first through that post on this link and then come back to try out the exercises.
Before we start, please allow me to give you some update about the blog. It is with an honour that I am announcing that MIB has been ranked among the top 40 Machine Learning blog to follow in 2019 alongside some very known Machine Learning blog like Google Machine Learning News, MIT News, Machine Learning Mastery and many more great ..read more

MIB | Machine Intelligence Blog

5y ago

One of the most used scientific computing library for python is without a doubt NumPy, Numpy is very fast at computing arrays since it is mostly written in C programming. NumPy adds support for large, multi-dimensional arrays and matrices, along with an extensive collection of high-level mathematical functions (for linear algebra) to operate on these arrays. Pandas, which is a library for data manipulation, is written based on NumPy. In this post, we will discuss NumPy and how to use it, and by the end of the post, you will see why it is one of the most famous Python libraries for data science ..read more