prakhr / coursera-specializations

Contains courses in specializations of coursera on deep learning

Coursera-Specializations on CNNs

[Coursera Specializations] - Python-Javascript-Programming-Tensorflow-Algorithms-Deployment Simple predictions and algorithms implemented with tensorflow and google colab(courses has the colab environment setup)
You can download 200 OK! from web apps and other project material from here

Table of contents

• Learnt Skills
• Setup
• Modules
• Certificate

---

Learnt

You will learn how to use different methods of tensorflow, in other words to build a pipeline from preprocessing steps to building models and finally saving models in a necessary format. You will also use javascript modules to read from csv files and predict outputs.

Then you will learn what are the necessary layers to put in different courses while building CNNs from scratch.

Apart from that you will learn hypertuning a parameter, how tweaking them affects your accuracy and convergence as well as instability.

---

Setup

The Sublime Text for writing javascript Environment download, documentation, and programming resources are available at: https://www.sublimetext.com/3.

I prefer setting up a anaconda environment, this way you can install different versions of tensorflow easily.

Or learn by running scripts on google colab using their sweet gpu for increased runtime(Note that your variables and files will get lost after a fresh run in colab).

---

Modules

• tensorflow
• keras
• shutil
• zipfile
• random
• google.colab
• numpy
• json
• matplotlib

---

Certificates

you can get a certificate like this if not choose to audit the course