neural networks and deep learning coursera week 4 programming assignment

FAQ: Frequently Asked Questions for all DLS Courses

Here is a list of Frequently Asked Questions about the DLS Courses, the assignments and other course related topics.

• Get a clean copy of an assignment
• All my previous work has disappeared!
• Name errors on predefined functions
• When to use np.multiply or np.dot
• Missing “Submit Assignment” button
• Source code for utility functions
• Access to assignments after completing the courses
• Managing course deadlines
• ML/DL textbooks
• Derivation of back propagation
• Symmetry Breaking
• Formatting mathematical formulas
• Adding cells to a notebook
• Coursera Help Center
• Move your topic to the right category
• Grader errors

Is there a way to restore one of my assignment notebooks to its original state?

Yes, please see this topic . Please note that the same procedure is applicable if you want to make sure that you have the latest version of any of the notebooks or supporting files.

Return to topics list.

Help! All my previous work has disappeared from one of my notebooks!

That usually means the course staff has published a new version of that notebook with some improvements. In most cases, your work has been saved and you can recover it. From the updated version of the notebook, click “ File → Open ” which gives you a “file navigation” view of the assignment. You should see another version of the notebook that has the date and time interpolated into the name: that’s the previous version containing your saved work. Click that one to open it. Then go back to the Programming Assignment tab and click Work in Browser again. Now you’ve got two parallel browser tabs showing your previous worked notebook and the new clean one. Carefully go through and “Copy/Paste” your completed code from the previous version to the new version. It’s important that you do that so that you are using the latest version of the notebook with the fixes that the course staff just published. Make sure to click “Save” when you’re finished copying everything and you should be all converted over to the new version.

But please note that sometimes the assignment updates that are extensive enough that previous student work does not carry over. In those cases, the only way to recover your work is to have manually saved a local copy of any completed notebooks. Saving your work locally is highly recommended just on general principles. There is a Reading Item in Course 1 Week 2 that shows how to do that. Here is a thread that shows how to save all the files associated with an assignment, not just the notebook itself.

NameError: name ‘sigmoid_derivative’ is not defined

If you get an undefined error for a function or variable defined earlier in the notebook, it means you need to execute the earlier cells first. Try “ Cell → Run All Above ” and then rerun the failing cell. Note that you need to run all the cells again every time you close and reopen one of the notebooks or restart the “kernel”.

Also note that just typing new code into a function cell and then calling the function again does nothing: it runs the old code again. You need to click “Shift-Enter” on the cell with the new code in order for the new version of the code to be loaded into the runtime image of the notebook.

These and other points about how the Jupyter Notebooks work are covered in some detail in the lecture called “A Quick Tour of the Jupyter Notebooks” in DLS C1 W2. There is a similar tour of the notebooks in Week 1 of MLS C1. If you skipped that video the first time through, it’s worth a look.

How do I know whether to use np.multiply or np.dot ?

Here’s a thread which discusses that. Note that Prof Ng will always use “*” when he means “elementwise” multiply. In numpy “*” and np.multiply are two different ways to write elementwise multiply.

There is no “Submit Assignment” button in my notebook

Note that the “Submit Assignment” button will usually not show up if you have renamed the notebook. Also note that if you do see the button in a version with a different name, clicking it submits the “official” notebook (the one opened by the “Work in Browser” or “Launch Lab” link), not the one you are currently running.

If you are running the graded notebook and don’t see the button, the first thing to try is “ Kernel → Restart and Clear Output ”. If that doesn’t restore the “Submit Assignment”, then also try closing and reopening the notebook. Try using a different computer or moving to a different network, especially if you are on a work or school network that may have restrictive IT policies.

This can also be caused by browser issues. If you are using a browser other than Chrome, please try switching to Chrome. Also try using an “incognito” window and clearing your browser cache. If that still doesn’t work, try the “Get a fresh copy” procedure earlier on this thread and then “copy/paste” your solutions over to the clean copy.

If none of these suggestions work, then please create a thread about this and it may be necessary to get the course staff or Coursera to take a look at your situation.

How can I find the source for utility functions invoked by the notebook?

Click " File → Open " and that will give you a file navigation view of the files for the current notebook. You can tell the file names to look for by examining the “import” cell early in the notebook. For example, the functions used to test your code in the notebook are usually in a file called public_tests.py .

Will I be able to access my completed assignments after I finish the courses?

Once you have successfully completed the specialization, your payment will automatically stop and after a few weeks you will also automatically lose access to the course assignments. The exact timing depends on the monthly renewal date for your subscription, but there is apparently no way to stop this from happening. It is also important to keep in mind that there is no option to re-enroll or continue payment for the course after completion. If you would like to continue to have access to the assignments after completion, you need to download all of the notebooks and associated files to your local computer before you lose access. The safest thing to do is to download each assignment once you’ve completed it (see below for instructions). That also gives you a backup copy in case there are any server problems. Note that the lectures and the Discourse forum discussions remain available even after you lose access to the assignments.

Here is a thread about how to download all the files associated with one of the assignments. There is no method for downloading all the assignments of a given course in one shot, so you’ll need to do each one individually.

Is it possible to reset the course deadlines?

Yes, the deadlines here are just “advisory” and there is no penalty for missing them. Just work at whatever pace makes sense for you given your other time commitments. If you hit one of the deadlines, you’ll get a notification and can use the “Push End Date” option as shown to reset to a later deadline with no effect on your grades:

Are there recommended textbooks that cover the material in this course?

Here’s a thread with links to various ML/DL text books.

Where can I find more information about the derivation of the back propagation formulas?

Prof Ng has specifically designed this course so that it does not require knowledge of matrix calculus in order to follow the material. Of course that also means we have to take a lot of the formulas on faith and don’t get to see why they turn out the way they do. If you have the math background and want to dig deeper, here’s a thread with some links to get started on that.

Why is Symmetry Breaking Required?

In Week 2 of Course 1, we successfully train a Logistic Regression model by starting with all zeros for the w and b weights and bias values. But when we get to Week 3, Prof Ng tells us we need to use random initialization of the parameters instead of using all zeros. Why does using all zeros work for Logistic Regression, but not for real Neural Networks? Please see this thread for a detailed discussion of this question.

How can I format mathematical formulas in a Discourse post?

Discourse supports LaTeX rendering by using the MathJax plugin. Just enclose your LaTeX expression with single dollar signs on each side. For example:

sin^2x + cos^2x = 1

was produced by entering the following without the backslashes:

\$sin^2x + cos^2x = 1\$

Can I add my own cells to a notebook?

Yes, you can use “Insert → Cell Below” to do that. Note that this could confuse the grader in some cases, so the best practice is to add this comment as the first line in any added cells:

How do I contact Coursera when I can’t find the answer I need in the Help Center?

First go to this page and use the search bar in the top corner to search for discussions relevant to your questions.

If you can’t log in or if chat isn’t available, fill in this form to contact them by email.

I accidentally posted my topic in the wrong category, how can I move it to the right category?

Be sure to save your changes by clicking the “checkmark” button when you are finished.

Grader errors - what do do when the grader does not give the full score

Expected value...

This can be caused if the notebook output cells contain large amounts of data, like an audio recording or generated graphics.

First check that your code passes all of the notebook tests and matches all of the expected results. If all the tests pass, then please try this sequence of commands:

The grader does not need to see your generated output: it only needs to call your functions. The above sequence will minimize the size of your notebook which can also help when the servers are busy.

Grader Error: Grader feedback not found

This error can also be caused by the notebook image being too large and containing syntax that confuses the grader. The first thing to try is the three steps shown above start with Kernel -> Restart & Clear Output . If that fails, then it may be that some of the “metadata” in your notebook is damaged. The next step is to get a clean copy of the notebook as shown in the first FAQ entry and then carefully “copy/paste” over your completed code into the clean copy and try submitting again.

Comment line with index: UNQ_C3 wasn't found in code

This error is caused by removing one of the comment lines given in the template function code cells. The grader uses those comment “tags” to identify the graded cells in some of the notebooks. Please be careful when modifying anything outside the “YOUR CODE HERE” segments in the template code in the notebooks. It’s not illegal to add or change things outside those bounds, but you need to be careful and understand the parts that the grader may use. You can recover lost or modified lines by getting a clean copy of the notebook as described here .

Keywords: FAQ

Deep-Learning-Specialization

Coursera deep learning specialization, sequence models.

This course will teach you how to build models for natural language, audio, and other sequence data. Thanks to deep learning, sequence algorithms are working far better than just two years ago, and this is enabling numerous exciting applications in speech recognition, music synthesis, chatbots, machine translation, natural language understanding, and many others.

• Understand how to build and train Recurrent Neural Networks (RNNs), and commonly-used variants such as GRUs and LSTMs.
• Be able to apply sequence models to natural language problems, including text synthesis.
• Be able to apply sequence models to audio applications, including speech recognition and music synthesis.

Week 1: Sequence Models

Learn about recurrent neural networks. This type of model has been proven to perform extremely well on temporal data. It has several variants including LSTMs, GRUs and Bidirectional RNNs, which you are going to learn about in this section.

Assignment of Week 1

• Quiz 1: Recurrent Neural Networks
• Programming Assignment: Building a recurrent neural network - step by step
• Programming Assignment: Dinosaur Island - Character-Level Language Modeling
• Programming Assignment: Jazz improvisation with LSTM

Week 2: Natural Language Processing & Word Embeddings

Natural language processing with deep learning is an important combination. Using word vector representations and embedding layers you can train recurrent neural networks with outstanding performances in a wide variety of industries. Examples of applications are sentiment analysis, named entity recognition and machine translation.

Assignment of Week 2

• Quiz 2: Natural Language Processing & Word Embeddings
• Programming Assignment: Operations on word vectors - Debiasing
• Programming Assignment: Emojify

Week 3: Sequence models & Attention mechanism

Sequence models can be augmented using an attention mechanism. This algorithm will help your model understand where it should focus its attention given a sequence of inputs. This week, you will also learn about speech recognition and how to deal with audio data.

Assignment of Week 3

• Quiz 3: Sequence models & Attention mechanism
• Programming Assignment: Neural Machine Translation with Attention
• Programming Assignment: Trigger word detection

Course Certificate

• 🌐 All Sites
• _APDaga DumpBox
• _APDaga Tech
• _APDaga Invest
• _APDaga Videos
• 🗃️ Categories
• _Free Tutorials
• __Python (A to Z)
• __Internet of Things
• __Coursera (ML/DL)
• __HackerRank (SQL)
• __Interview Q&A
• _Artificial Intelligence
• __Machine Learning
• __Deep Learning
• _Internet of Things
• __Raspberry Pi
• __Coursera MCQs
• __Linkedin MCQs
• __Celonis MCQs
• _Handwriting Analysis
• __Graphology
• _Investment Ideas
• _Open Diary
• _Troubleshoots
• _Freescale/NXP
• 📣 Mega Menu
• _Logo Maker
• _Youtube Tumbnail Downloader
• 🕸️ Sitemap

Coursera: Neural Networks and Deep Learning (Week 4B) [Assignment Solution] - deeplearning.ai

▸  deep neural network for image classification: application. i have recently completed the neural networks and deep learning course from coursera by deeplearning.ai, while doing the course we have to go through various quiz and assignments in python. here, i am sharing my solutions for the weekly assignments throughout the course. these solutions are for reference only. >  it is recommended that you should solve the assignments by yourself honestly then only it makes sense to complete the course. >  but, in case you stuck in between, feel free to refer to the solutions provided by me., don't just copy paste the code for the sake of completion.  even if you copy the code, make sure you understand the code first. click here : coursera: neural networks & deep learning (week 4a) scroll down  for coursera: neural networks & deep learning (week 4b) assignments. (adsbygoogle = window.adsbygoogle || []).push({});.

Recommended Machine Learning Courses: Coursera: Machine Learning    Coursera: Deep Learning Specialization Coursera: Machine Learning with Python Coursera: Advanced Machine Learning Specialization Udemy: Machine Learning LinkedIn: Machine Learning Eduonix: Machine Learning edX: Machine Learning Fast.ai: Introduction to Machine Learning for Coders

Deep Neural Network for Image Classification: Application

When you finish this, you will have finished the last programming assignment of week 4, and also the last programming assignment of this course you will use use the functions you'd implemented in the previous assignment to build a deep network, and apply it to cat vs non-cat classification. hopefully, you will see an improvement in accuracy relative to your previous logistic regression implementation. after this assignment you will be able to: build and apply a deep neural network to supervised learning. let's get started, 1 - packages.

• numpy  is the fundamental package for scientific computing with Python.
• matplotlib  is a library to plot graphs in Python.
• h5py  is a common package to interact with a dataset that is stored on an H5 file.
• PIL  and  scipy  are used here to test your model with your own picture at the end.
• dnn_app_utils provides the functions implemented in the "Building your Deep Neural Network: Step by Step" assignment to this notebook.
• np.random.seed(1) is used to keep all the random function calls consistent. It will help us grade your work.

2 - Dataset

Check-out our free tutorials on IOT (Internet of Things):

3 - Architecture of your model

• A 2-layer neural network
• An L-layer deep neural network

3.1 - 2-layer neural network

• The input is a (64,64,3) image which is flattened to a vector of size  ( 12288 , 1 ) .
• The corresponding vector:  [ x 0 , x 1 , . . . , x 12287 ] T  is then multiplied by the weight matrix  W [ 1 ]  of size  ( n [ 1 ] , 12288 ) .
• You then add a bias term and take its relu to get the following vector:  [ a [ 1 ] 0 , a [ 1 ] 1 , . . . , a [ 1 ] n [ 1 ] − 1 ] T .
• You then repeat the same process.
• You multiply the resulting vector by  W [ 2 ]  and add your intercept (bias).
• Finally, you take the sigmoid of the result. If it is greater than 0.5, you classify it to be a cat.

3.2 - L-layer deep neural network

• The input is a (64,64,3) image which is flattened to a vector of size (12288,1).
• The corresponding vector:  [ x 0 , x 1 , . . . , x 12287 ] T  is then multiplied by the weight matrix  W [ 1 ] W [ 1 ]  and then you add the intercept  b [ 1 ] . The result is called the linear unit.
• Next, you take the relu of the linear unit. This process could be repeated several times for each  ( W [ l ] , b [ l ] )  depending on the model architecture.
• Finally, you take the sigmoid of the final linear unit. If it is greater than 0.5, you classify it to be a cat.

3.3 - General methodology

4 - two-layer neural network.

5 - L-layer Neural Network

6) Results Analysis

• Cat body in an unusual position
• Cat appears against a background of a similar color
• Unusual cat color and species
• Camera Angle
• Brightness of the picture
• Scale variation (cat is very large or small in image)

7) Test with your own image (optional/ungraded exercise)

• for auto-reloading external module:  http://stackoverflow.com/questions/1907993/autoreload-of-modules-in-ipython
• coursera.org

Hi sir , in week 4 assignment at 2 layer model I am getting an error as" cost not defined"and my code is looks pretty same as the one you have posted please can you tell me what's wrong in my code

yes even for me .. please suggest something what to do

Have you tried running all the cell in proper given sequence. Because, In jupyter notebook a particular cell might be dependent on previous cell. I think, there in no problem in code. You are doing something wrong with the executing the code. Please check once.

fundamentals of scalable data science week 1 assignment in coursera solution I am finding some problem

Hi. i seen function predict(), but the articles not mention, thank sir

hi , im getting a " too many values to unpack : error and ive used your code only kindly help me get this

for which cell you are getting this error??

brother can u kindly whatsapp me on 9735255861 ? then i can share with u some screenshots of the error that we are facing. The old code is no more running without error in th week 4 assignments. kindly help me out . we cannot even pass the assignments. :|

Hi, In case the assignment question is changed for week 4 then I am now aware of that. But the question is same then I am sure about the solutions. I don't have access to the course now. So can't check it now.

Our website uses cookies to improve your experience. Learn more

Contact form