deep learning specialization assignments github

Deep-Learning-Specialization-Coursera

This repo contains the updated version of all the assignments/labs (done by me) of deep learning specialization on coursera by andrew ng. it includes building various deep learning models from scratch and implementing them for object detection, facial recognition, autonomous driving, neural machine translation, trigger word detection, etc., deep learning specialization coursera [updated version 2021].

Announcement

[!IMPORTANT] Check our latest paper (accepted in ICDAR’23) on Urdu OCR

This repo contains all of the solved assignments of Coursera’s most famous Deep Learning Specialization of 5 courses offered by deeplearning.ai

Instructor: Prof. Andrew Ng

This Specialization was updated in April 2021 to include developments in deep learning and programming frameworks. One of the most major changes was shifting from Tensorflow 1 to Tensorflow 2. Also, new materials were added. However, Most of the old online repositories still don’t have old codes. This repo contains updated versions of the assignments. Happy Learning :)

Programming Assignments

Course 1: Neural Networks and Deep Learning

• W2A1 - Logistic Regression with a Neural Network mindset
• W2A2 - Python Basics with Numpy
• W3A1 - Planar data classification with one hidden layer
• W3A1 - Building your Deep Neural Network: Step by Step¶
• W3A2 - Deep Neural Network for Image Classification: Application

Course 2: Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization

• W1A1 - Initialization
• W1A2 - Regularization
• W1A3 - Gradient Checking
• W2A1 - Optimization Methods
• W3A1 - Introduction to TensorFlow

Course 3: Structuring Machine Learning Projects

• There were no programming assignments in this course. It was completely thoeretical.
• Here is a link to the course

Course 4: Convolutional Neural Networks

• W1A1 - Convolutional Model: step by step
• W1A2 - Convolutional Model: application
• W2A1 - Residual Networks
• W2A2 - Transfer Learning with MobileNet
• W3A1 - Autonomous Driving - Car Detection
• W3A2 - Image Segmentation - U-net
• W4A1 - Face Recognition
• W4A2 - Neural Style transfer

Course 5: Sequence Models

• W1A1 - Building a Recurrent Neural Network - Step by Step
• W1A2 - Character level language model - Dinosaurus land
• W1A3 - Improvise A Jazz Solo with an LSTM Network
• W2A1 - Operations on word vectors
• W2A2 - Emojify
• W3A1 - Neural Machine Translation With Attention
• W3A2 - Trigger Word Detection
• W4A1 - Transformer Network
• W4A2 - Named Entity Recognition - Transformer Application
• W4A3 - Extractive Question Answering - Transformer Application

I’ve uploaded these solutions here, only for being used as a help by those who get stuck somewhere. It may help them to save some time. I strongly recommend everyone to not directly copy any part of the code (from here or anywhere else) while doing the assignments of this specialization. The assignments are fairly easy and one learns a great deal of things upon doing these. Thanks to the deeplearning.ai team for giving this treasure to us.

Connect with me

Name: Abdur Rahman

Institution: Indian Institute of Technology Delhi

Find me on:

deep-learning-specialization

Deep learning specialization.

These are my solutions for the exercises in the Deep Learning Specialization offered by Andrew Ng on Coursera.

• Logistic Regression with a Neural Network mindset
• Planar data classification with one hidden layer
• Building your Deep Neural Network - Step by Step
• Deep Neural Network - Application
• Structuring Machine Learning Projects
• Sequence Models

All the images/notebooks shown here have been taken from the Deep Learning specialization on the Coursera platform. They are shown here just for educational purposes.

Deep-Learning-Specialization

Coursera deep learning specialization, convolutional neural networks.

This course will teach you how to build convolutional neural networks and apply it to image data. Thanks to deep learning, computer vision is working far better than just two years ago, and this is enabling numerous exciting applications ranging from safe autonomous driving, to accurate face recognition, to automatic reading of radiology images.

• Understand how to build a convolutional neural network, including recent variations such as residual networks.
• Know how to apply convolutional networks to visual detection and recognition tasks.
• Know to use neural style transfer to generate art.
• Be able to apply these algorithms to a variety of image, video, and other 2D or 3D data.

Week 1: Foundations of Convolutional Neural Networks

Key concepts of week 1.

• Understand the convolution operation
• Understand the pooling operation
• Remember the vocabulary used in convolutional neural network (padding, stride, filter, …)
• Build a convolutional neural network for image multi-class classification

Assignment of Week 1

• Quiz 1: The basics of ConvNets
• Programming Assignment: Convolutional Model: step by step
• Programming Assignment: Convolutional Model: application

Week 2: Deep convolutional models

Key concepts of week 2.

• Understand multiple foundational papers of convolutional neural networks
• Analyze the dimensionality reduction of a volume in a very deep network
• Understand and Implement a Residual network
• Build a deep neural network using Keras
• Implement a skip-connection in your network
• Clone a repository from github and use transfer learning

Assignment of Week 2

• Quiz 2: Deep convolutional models
• Programming Assignment: Residual Networks

Week 3: Convolutional Neural Networks

Key concepts of week 3.

• Understand the challenges of Object Localization, Object Detection and Landmark Finding
• Understand and implement non-max suppression
• Understand and implement intersection over union
• Understand how we label a dataset for an object detection application
• Remember the vocabulary of object detection (landmark, anchor, bounding box, grid, …)

Assignment of Week 3

• Quiz 3: Detection algorithms
• Programming Assignment: Car detection with YOLO

Week 4: Special applications: Face recognition & Neural style transfer

Discover how CNNs can be applied to multiple fields, including art generation and face recognition. Implement your own algorithm to generate art and recognize faces!

Assignment of Week 4

• Quiz 4: Special applications: Face recognition & Neural style transfer
• Programming Assignment: Art generation with Neural Style Transfer
• Programming Assignment: Face Recognition

Course Certificate

Coursera Deep Learning Specialization; thoughts and tips

My thoughts (and tips) on the coursera 5-course deep learning specialization. ¶.

I recently completed the Deep Learning specialization (a 5-course sequence) on the Coursera platform which was developed by deeplearning.ai with Andrew Ng. I had an overall very positive experience with it, and felt like it was well worth the cost to have Andrew Ng personally curate, organize and explain the most important ideas in the field. My course notes and some annotated code from the homeworks are available in this repo .

My background going into the course was that I had taken the Stanford University Machine Learning course on Coursera, and then actually used some machine learning on a handful of personal projects as well as in a one-year, part-time data scientist role. I also had a strong quantitative and python programming background (including lots of the python ecosystem for data science, but not tensorflow or keras).

TL;DR ¶

• Well worth the money; it's like paying Andrew Ng $50 a month to personally curate all the DL knowledge for you.
• Homeworks spoon-feed you code; expect to append your own project(s) to the specialization to really practice your implementation skills.
• Keep a course notebook; be diligent about it, force yourself to be concise, add frequent summary sections where you review the content.
• Invest some time in working through a good external resource for learning tensorflow and keras and you will get much more from the homeworks in the later courses.
• Don't forget to download your full jupyter workspace for each class if you want to be able to access and run your homework notebooks locally after your subscription is over!

My Timeline ¶

I worked through the specialization in about 3.5 weeks working on it around 30 hours per week, which meant I only needed to subscribe to the specialization for a single month ($50 total). Each course is organized by week, and on a full day of work I could go through one week's lecture content and the weekly quiz in the morning and the corresponding homework(s) in the afternoon. Because I find that Andrew Ng speaks a bit slowly and because I have a solid math and programming background, I tended to watch the lectures on 1.75x speed (you can work up to this over the course of a couple videos). However I also paused the lectures a lot to take notes and review ideas, so on average I was spending more time on each week's content than the estimated amounts listed on the course websites.

My Approach ¶

For each course I created a dedicated jupyter notebook for course notes, organized by week and weekly sub-topics (these notebooks are available on github here ). I would pause throughout each lecture video to add concise notes to my notebook, including screen captures of the most important slides. I often had the experience of beginning to type up some notes, only to realize there was some gap in my understanding which prompted me to skip back in the video to clarify; without taking notes I wouldn't even have been aware of my own misunderstandings. Whenever it felt like we were moving into a new topic I would add a summary section to the notebook where I would read back through my notes and try to summarize the content as concisely as possible. I highly recommend this approach for these reasons:

• The notes are extremely helpful for both the quiz and homeworks.
• Concise notes give me a very easy way to refer back to ideas (and review important slides and figures) after the course is over.
• Typing up notes as you go solidifies your learning and highlights any gaps in your understanding.
• Typing up the summary section gives you a built in review of the content to help solidify the ideas.

Below is a screen cap of an example section from my CNN course notebook; the short summary section is in purple and yellow highlighting at the bottom.

The homeworks are in the form of jupyter notebooks served out from a remote server, which have a submission button for grading. At the end of each course I looked back through the homework notebooks and pulled out any functions and code snippets that I wanted to be able to reference quickly, and also typed up summaries of how the code was organized in the homeworks, as I found this very helpful for solidifying the ideas. Many of the homework notebooks are behind a paywall, so I also made sure to download the full jupyter workspace for the course (not just the homeworks, but data and helper scripts also) so I could rerun the homework locally after my subscription ended. Instructions for that are here .

My Thoughts ¶

Mathematical Difficulty. Similar to the Stanford University course, Andrew Ng does a great job simplifying (or downright ignoring) the more detailed or difficult mathematical pieces of the content. This lets you develop some mastery of broad ideas and guiding principles without getting too bogged down; if you want the math afterwards just pick up any textbook. All the mathematical notation conventions for referring to different components of the neural networks can be a bit overwhelming, so make sure to review them often or make yourself a cheat sheet as needed.

Programming Difficulty. The programming assignments very much spoon-feed you the code. But they are still helpful in demonstrating not just the sematics of implementing various things, but also how best to functionalize/organize various higher-level tasks like training and evaluating a model. If you are fluent with general python programming and numpy you will find many of the programming assignments very easy. One area that is VERY lacking is a solid introduction to tensorflow and keras , and it would definitely be worthwhile to spend a day going through an external resource for this. The mechanics of both of these are a bit counterintuitive / nontrivial, and once the specialization switches into using them for homework (at the end of the second course) they assume a certain amount of fluency. I often had the experience of being able to complete a homework because of the level of spoon-feeding, but not really feeling like I understood what my TF or keras code was doing.

Preparation for DL Projects. Because the homeworks do not really give you a good grasp of tensorflow or keras and spoon feed you so much, you really are not in a position to confidently fire up your -> insert favorite python front end <- and implement a DL project after completing this specialization. I would plan on adding a self-directed project (kaggle competition, whatever) to the end of the specialization to round out your actual DL implementation skills. For instance I just joined a Kaggle competition for disease classification from retina images, where I'll use CNNs.

Lecture Quality. You can't beat Andrew Ng for giving concise, intuitive explanations of concepts in ML. I had basically zero exposure to DL prior o the specialization, but I do feel like the specialization touched on all the common DL "buzzwords" I had previously encountered. The editing in many of the videos (especially in the last two courses) leaves something to be desired, but I didn't think this really detracted from the content. I think there is a lot to be said for Andrew's approach of hand-drawing figures/diagrams/pictures on the fly, but I sometimes wished he would have followed this up with a slide with a cleaner version of the figure.

Breadth vs. Depth. Hands down this specialization gives you breadth over depth; assuming you master the material, you will have a general idea of the approach for the large majority of influential DL architectures and use cases. There is no serious depth on any one topic, but there is sufficient coverage that you are well prepared to go deeper (hello, InceptionNet) on your own.

Technical Bugs. Expect to deal with a few headaches around the homework notebook technology (you are missing files in your workspace on the server, the grading process is timing out on your notebook submission, etc.). Annoyingly, the homework notebook stuff is provided not by Coursera but by an unspecified third party, so Coursera will refuse to help you troubleshoot any issues. The only way to deal with these bugs is exhaustive search in the course forums for threads dealing with similar issues. More than once I spent upwards of an hour trying to solve an issue, which was pretty frustrating.

Worth the Money? In my opinion, definitely. It is certainly true that all of the content and ideas in the course are freely available online in the form of blog posts, articles, stack exchange threads, github repos etc. But like Linux, that content is only free if your time is worth nothing :) You should think of the subscription fee as paying an absurdly low price ($50 / month) to have Andrew Ng personally find, curate and organize all this material (and then ELIF it to you).

• Terms and Conditions
• Cookie Policy
• Privacy Policy

• 10×10 Puzzle - A reincarnation of the classic 10×10 puzzle game, but with some twists! Fill rows and columns with blocks to clear them,...
• Road Fixer - Engaging puzzle game that challenges players to clear obstacles and pave the way for a car to successfully navigate to the...
• Christmas N Tiles - Christmas N Tiles is a fun and simple elimination mahjong-inspired game, your goal is to move tiles onto the bottom of the...
• Parkour Blockcraft - Welcome to the world of Blockcraft! Jump over the blocks to reach the portals! Be careful you may fall down. Finish the game...
• Princess Get Ready For Winter - Welcome to the Princess Get Ready For Winter. Winter is almost there! Princess Anna, Elsa, Moana, Ariel, Jasmine, and Snow...
• Rainbow Girls Perfect Winter Outfits - Welcome to Rainbow Girls Perfect Winter Outfits. Winter is not a season, It’s a celebration. Enjoy this winter with...
• Relax Bubble Shooter - Relax Bubble Shooter is a fantastic game suitable for the whole family and children, combining relaxation and mental challenge....
• Amazing Digital Circus Pomni - Amazing Digital Circus – Pomni is a vibrant coloring game that will captivate you with its interactive coloring pages...
• Brain Test Tricky Puzzles - Brain Test Tricky Puzzles is a fun and smart 2D quiz puzzle game for kids. You have to solve different but simple questions...
• World of Alice Sizes - World of Alice – Sizes World of Alice – Sizes is a game developed for children to teach the difference in size...

Digital Circus Run and Shoot is an exhilarating and fast-paced mobile game that combines the excitement of a ...

Digital Circus Town Builder is a captivating simulation game that combines the charm of a circus with the ...

Do it up! - and nothing else! If, at times, you don’t understand what’s going on around you, then here ...

City Idle Counter Terrorists is an exhilarating mobile game that combines strategy and action. In this game, you ...

Welcome to the world of Blockcraft! Jump over the blocks to reach the portals! Be careful you may fall down. ...

In Blocky Universe, you set out on an adventure in a world full of monsters! With your trusty bow and your ax, its ...

A great Snake IO game is beginning with the Worms Arena IO game! The baits are all around the map and collect them ...

StickMan Stunt Race 3D is an action-packed game that combines parkour and racing in a fun and exciting way. You ...

Amazing Digital Circus - Pomni is a vibrant coloring game that will captivate you with its interactive coloring ...

Embark on a charming adventure with Relaxing BananaCAT Clicker, an engaging mobile game that combines feline ...

Rev up your engines and brace yourself for the ultimate gridlock showdown in Traffic Jam 3D! Take on the role of a ...

Blast off into an intergalactic adventure with Space Mission Truck, the high-octane game that puts you in the ...

Embark on an exciting mining adventure with Amazing Gold Miner! Dig deep into the earth, navigate treacherous ...

In Pixel Demolisher Cannon, we destroy various objects consisting of pixel maps. Shoot bombs at objects and make ...

Welcome to Xmas Breakout game! Move left and right to control the ball and start beaking christmas items. Clear ...

The aim of the game is simple - press the button. The more times you click on the button, the more points you earn ...

Welcome to the Princess Get Ready For Winter. Winter is almost there! Princess Anna, Elsa, Moana, Ariel, Jasmine, ...

Welcome to Rainbow Girls Perfect Winter Outfits. Winter is not a season, It’s a celebration. Enjoy this ...

Right after midnight, the most famous K-pop band of the moment will entertain the gathering with their cool songs, ...

Christmas Coloring Game 2 is a very popular drawing game. As you can see from the name, this game is full of ...

Navigate through the levels by extending the limbs of the character while steering clear of obstacles. Be cautious ...

Single player infinity Run Funny Arcade Stickman Parkour Game by Baran Yel Super Cluster Hub . Coming Soon; Market ...

Welcome to stickman base defense - the ultimate arcade idler browser game! Immerse yourself in the world of ...

Welcome to the incredible monster hero fighting city mission of 2021. The concept of this game very unique in ...

Would You Rather is the ultimate group game to play with friends! This party app contains over 120 of the best ...

In this new adventure, you must reach the portal. To reach the portal, you should collaborate with your friend and ...

welcome to Rolling Balls io game! !use the joystick or mouse click to control you ball and start to crush ...

Play one of the most exciting holdem games in the world. Pick your avatar and Win at 2 types of table room you ...

Relax Bubble Shooter is a fantastic game suitable for the whole family and children, combining relaxation and ...

Farm Life is a game that simulate tasks of farmers who work on a big farm. With farm Life, Players are transformed ...

Little Mermaid Endless Runner, you play as the Little Mermaid, as she swims through the ocean collecting pearls. ...

Lets splash some dirt over the walking people near the road with this Dirty Them All game! Drive your car through ...

A reincarnation of the classic 10x10 puzzle game, but with some twists! Fill rows and columns with blocks to clear ...

Engaging puzzle game that challenges players to clear obstacles and pave the way for a car to successfully ...

Christmas N Tiles is a fun and simple elimination mahjong-inspired game, your goal is to move tiles onto the ...

Brain Test Tricky Puzzles is a fun and smart 2D quiz puzzle game for kids. You have to solve different but simple ...

Welcome to our newest Christmas theme game Santa Stars Shooter. The game has 20 levels with different difficulty. ...

Cameraman vs Skibidi Monster is an action-packed cameraman game where you engage in thrilling battles against ...

Tower Defense Kingdoms is an exciting and challenging game that puts you in control of defending your kingdom ...

This is a demo of the game Operation Thunderstrike. The game has infleunces from classic titles like Metal Slug, ...

Take part in an amazing tennis match and enjoy the swift and fluent game. Practice before joining a tournament, ...

Get your ride and explore aTopBike Racing & Moto 3D Bike 2023 ! Compete against other players around the world ...

You can play friendly match or league match in this game. Heading - Super League: Control the ball with your head, ...

Control a monster truck and pass many challenging obstacles in Hard Wheels Winter 2. Get to the finish without ...

Manage your ant colony, collect resources and increase the army of ants. Capture other insects or fight against ...

This game is in beta, you are welcome to try it but some bugs are to be expected.You get 5 gems each round. Keep ...

“Air Strike” is a very fun and colorful helicopter game in cartoon style! The game takes place in an enemy ...

Welcome to the exciting world of "AI anime streamers"! This is an exciting clicker game that will take you into ...

PRESS ENTER TO SEARCH

• WordPress.org
• Documentation
• Learn WordPress

Regarding Uploading course assignments on Github

I want to ask that can I upload the weekly assignments of this course on my Github account?

Hi @sengourav012

Welcome to the community.

Regards your question sharing your assignments it is a violation of Deeplearning.ai Policy:

You may not share your solutions to homework, quizzes, or exams with anyone other than a DeepLearning.AI mentor or staff member. This includes posting solutions to Discourse , Github, or any other code repository. If you are found doing this, DeepLearning.AI reserves the right to take action against you.

For more information regards deeplearning.ai code of conduct, please, take a look at Deeplearning.ai guideline

best regards elirod

Hi @sengourav012 ,

Please follow the guidelines @elirod shared with you.

However, I will like to elaborate on one point: You can upload your weekly assignments in a GitHub repo which is private and only visible to you . That is to say, you can keep them for your own personal use. You are not allowed to put them publicly or share them with others.

Best, Mubsi

Navigation Menu

Search code, repositories, users, issues, pull requests..., provide feedback.

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly.

To see all available qualifiers, see our documentation .

• Notifications You must be signed in to change notification settings

Pull requests: morriwong22/Coursera-Deep-Learning-Specialization-Assignments-and-Quiz

Welcome to pull requests.

Pull requests help you collaborate on code with other people. As pull requests are created, they’ll appear here in a searchable and filterable list. To get started, you should create a pull request .