Course notes of: AI for medical specialization
The best way to get deeper understanding of machine learning is to apply machine learning to different areas. One of the most exciting applications is the field of medicine. This course will focus on teaching how we can diagnose pneumonia from unstructured data or identify heart attack risk based on patient’s lab result.
It is a three course specialization.
Why would you be interested in reading my journey? I can give plenty of reasons but the one that comes to my mind right after asking this question is:
Those girls or women who are ambitious about their career, but also feel the importance of having a family, and often end up struggling in both should read it. Also, the men who want to understand their partner’s dilemma and the mixture of emotions while planning for one should definitely read it. Hence let’s dive into it right away.
Disclaimer: This article is trying to cover several scattered thoughts. It might…
So far we have learned to divide the set into train, development, and test set. We also learned of regularization. In this article, we will learn about ways to speed up the training of a neural network. Without any further delay let’s get started.
If you have not checked out my article on week 1 notes please quickly refer to this link.
While training a neural network, one of the techniques to speed up the training is to normalize inputs. For example, if we have two features in our training set then we can calculate the mean and variance for…
I started writing the course notes for the deep learning specialization course from Coursera. As I write, my concepts get clear and I get a better sense of the topic. I hope these notes are useful for you all.
With this article, I am going to jump to course 2 of the informative specialization course. If you have not checked out my articles on the first course please go through this link.
This course will focus on practical aspects of machine learning such as learning the aspects to make the neural network work well. It will range from hyperparameter tuning…
This week, we will learn about neural networks. If you have not gone through the previous week’s notes please refer to week 1 and week 2 notes.
We will start with the neural network with a single hidden layer.
X1, X2, X3 are input features stacked vertically, which is called an input layer. The middle layer with 4 circles is known as the hidden layer. The name hidden is given because the operations performed in this layer are not observed. The last circle is known as the output layer. Each layer is identified with a letter and a superscript number.
…
I started writing the notes for this particular course and this article contains week two notes. If you have not referred, week 1 notes please check out this article.
Binary Classification
Logistic Regression is an algorithm for binary classification. Let's take the problem of identifying cats with some input images. Here X would be some set of images and y would be an output label as 0 or 1. Further, let’s look at how the images are represented on a computer. They are stored as three separate channels of red, blue, and green color. So if the image is 64…
My Curiosity grew in AI when I first went through the machine learning course from Andrew Ng on Coursera. I was mesmerized by the capabilities of AI and the ways it can influence human lives. Knowingly or unknowingly AI touches several areas of our day today’s work, especially deep learning. It has transformed the traditional web search and made a greater impact in various domains such as computer vision, agriculture, healthcare, personalized education, autonomous driving, and many more.
Since then I decided to deep dive into this field which landed me in the 5 courses deep learning specialization from Coursera…
Words such as periods or menstruation, create discomfort if spoken loudly. Neither female nor their male counterpart feels comfortable talking about it. In districts, towns, and cities across India, menstruation remains a biological event shrouded in mystery and taboo, not to be spoken about openly. Myths about menstruation cause the girls to drop out of the schools or be ostracized during their menstrual days every month.
In this article, I would like to touch base on several topics related to menstruation and how advances in artificial intelligence can play a pivotal role to improve the lives of women and in…
This article contains notes of second week in the course of AI for medical diagnosis. If you have not gone through my first week’s notes, please checkout this article.
Without further delay, I would like to start with:
Model Evaluation
In order to decide, how good a model is, model evaluation is required. One of the important metrices for evaluation is Accuracy. When computing accuracy on a test set, we look at the proportion of total examples that the model correctly classified. Accuracy can be used further to derive evaluation metrices such as sensitivity and specificity.
Accuracy can be considered…
People form judgement about things, other people, rituals or culture quickly by using their primary thinking. Primary thinking is nothing but our instinct and it does not require intense focus. It helps people build experiences and as a person grows; experience and judgement together form perspectives.
These perspectives do not change until you keep an open mind i.e look at the other side of the coin.
One might wonder why it is important to come out of your comfort zone and pressurize the mind to learn something new or to develop a new perspective on your existing thinking?
I believe…
A fantasizer who wants to contribute towards a change..!!