What do the parameters in the Random Forest algorithm really mean?

About this article

In this article, we will try to get a deeper understanding of what each of the parameters does in the Random Forest algorithm. This is not an explanation of how the algorithm works. ( You might want to start with a simple explanation of how the algorithm works, found here in the link — A pictorial guide to understanding Random Forest Algorithm.)


The packages we will be looking at are


( for the Random Forest Classifier algorithm found in the sklearn library )


(for the Random Forest regressor algorithm)

Random Forest Classifier — parameters

  1. n_estimators ( default = 100 )

Since the RandomForest

An intuitive understanding of the torchvision library — basics to advanced ( Part 1/3 )

What is torchvision?

Torchvision is a library for Computer Vision that goes hand in hand with PyTorch. It has utilities for efficient Image and Video transformations, some commonly used pre-trained models, and some datasets ( torchvision does not come bundled with PyTorch, you will have to install it separately. )

About this article

This series of articles provides you with an understanding of what comprises the torchvision library — and also will look into how to implement these functionalities.

We will be using version 0.8.2 of torchvision

Part 1 gives you an over view of the features of torchvision

In Part 2 we will cover the…

A journey through the evolution of Intelligent Machines ( Part 1 – Early 40s till Y2K )

About this article

This is Part 1 of the two article series where we look into a brief history of events that give us an understanding about how AI as we know today has evolved. Part 1 takes us through the important events from 1940s till the late 1990s.

This article is NOT JUST about the history of machines that replaced humans for ‘mere’ computation or automation. ( If it were so, we would have called it — A brief history of Computers ). This is about devices or ideas that complemented or surpassed human intelligence in the real sense of ‘thinking’.


How to remove the HTML tags from your corpus for building your NLP data-set

Image source : Jackson So ( unsplash.com)

This article is part of the supporting material for the story — ‘Understanding NLP — from TF-IDF to transformers


Most of the times when you want to process a tonne of html files in your corpus, you would have to think about cleaning the HTML as a pre-processing step. Here are 3 ways to do the same.

1. Using Regex

Regular expressions are the most popular and powerful method for any of the complex string extraction process you want to carry out. …

A list of things-to-do for data pre-processing for creating Machine Learning data-sets ( and a few handy TIPS )

Image courtesy: unsplash.com

In this article

In this article, we will see what the data processing steps involved in pre-processing are, and some relevant codes in python to perform these actions.

We will also see the need to build an exhaustive check-list of pre-processing steps that you can apply on your data-set. A starter checklist is provided in this article. It can be used as a base to build on for specific projects that you are handling.


Most of the machine learning algorithms need clean datasets to be provided as input to the algorithm.

These are your train and test datasets . These datasets are further…

6 reasons why learning using kaggle competitions do not train you for the real world data science problems.

Image Courtesy : Ben Stern on unsplash.com

First things first — We have been on kaggle for over 4 years and would like to acknowledge the role Kaggle has played in our Data Science journey, as a very good learning platform for the beginner.

A little background — Recently we got the badge for becoming a competitionsKaggle Expert’ something that only a fraction of the users on the platform end up becoming. But Kaggle was a fleeting novelty. Soon after, we moved to a more realistic platform — the real world !!

7 use-cases where you can make your python code more nifty, concise, and elegant — without compromising readability.

Original image from unsplash.com ( https://unsplash.com/@thisisramiro )

The joy of coding Python should be in seeing short, concise, readable classes that express a lot of action in a small amount of clear code — not in reams of trivial code that bores the reader to death.”

- Guido van Rossum (creator of the Python programming language)

About this article

This article will cover 7 instances of code where you can convert your non-pythonic code into more legible and elegant code. Does your code have these usages ? Then, its a good time to revisit your old code and learn to write it in a more elegant manner.

The topics that…

The new AI features in Photoshop 2021 inspired by the latest researches in AI

About this article

Adobe Photoshop 2021 claims to be the ‘most advanced AI tool available for creatives’ so far. They have successfully adopted the latest research in AI and have added as tools in its latest version — Adobe Photoshop 2021.

Here we will take a look into the top 5 functionalities in the app that are facilitated by the advancements in AI and also have a look at the underlying AI research papers which we saw last year.

  1. Neural Filters
  2. Select Item
  3. Sky Replacement
  4. Content aware fill
  5. Discover panel and Quick actions

Neural Filters

With a bunch of filters added in the beta stage…

An intuitive understanding of the torchvision library — with 14 visual examples of transforms ( Part 2/3 )

For Part 1 ( introduction to the modules in torchvision ) , please visit the link below.

In this article

We will experiment with

  • some basic image transforms while loading a data-set into your PyTorch scripts

1. transforms

transforms are simple image transformation functions that can be carried out in a sequence soon after the dataset ( images) is loaded . These are the broad steps for performing image transformation using torchvision

  1. Define your custom transforms pipeline ( using torchvision.transforms.Compose )

( This just means , list down the different transformations to be done on your imageset )

2. Load your datasets and pass…

Pickling in machine learning — why and how to use it

Original Image Source (unsplash.com) Edited by Author

What is pickling?

The jargonized version“Pickling is the process of serialization and de-serialization of an object.”

The simplified versionLet’s just put it this way — pickling is a way of saving a python object on to your hard disc so that you can ‘unpickle’ it and use the object in another program / or transfer to another machine.

Now lets break down the jargons — in simple words ,

  • serialization means — writing from the memory to hard disc.
  • deserialization means — reading a file from hard disc on to your memory ( more sepcifically — accessible by the…


Making AI available to everyone. One commit at a time. A.I. M.L| | Quantitative programming | IoT | Python | Arduino | Comp Vision | contact@pythoslabs.com

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store