Lecture 2 - Working with Images & Logistic Regression

Livestream Link : https://youtu.be/4ZZrP68yXCI

Lecture Timing : May 30 (Sat), 8.30 AM PST/9:00 PM IST (2 hours)

Notebooks :

  1. Logistic Regression: https://jovian.ml/aakashns/03-logistic-regression
  2. Logistic Regression minimal starter: https://jovian.ml/aakashns/mnist-logistic-minimal
  3. Linear Regression minimal starter: https://jovian.ml/aakashns/housing-linear-minimal

Also see the notebooks from Lecture 1 for reference & revision.

What to do after the lecture?

Asking/Answering Questions :
Reply on this thread to ask questions during and after the lecture. Before asking, scroll through the thread and check if your question (or a similar one) is already present. If yes, just like it. During the lecture, we’ll answer 8-10 questions with the most likes. The rest will be answered on the forum. If you see a question you know the answer to, please post your answer as a reply to that question. Let’s help each other learn!

Attendance : NO NEED to mark attendance, we have removed the attendance requirement for the certificate!

39 Likes

Cant wait for Second Lecture today. @aakashns. All the best <3

9 Likes

Sir
Can I watch the lectures later if I have some network related issues because I prefer to watch after downloading as this assures continuous playback without delay. I will watch the live stream but I am asking in case if some issue occurs.

5 Likes

You have 24 hours until you watch the lecture. I think attendence isn’t compulsory now.

3 Likes

Hi!!
@aakashns would be great if we may share the codes and details a day before the session so that we may download and use the code in our local machines. Binder becomes unresponsive real time, so having some time would help!

Thanks!

9 Likes

Yes, you can watch it later. It is available on freecodecamp YouTube channel.

4 Likes

Question: I remember from college classes (we were doing numerical methods by hand or octave/matlab) that most iterative algorithms had a ‘stop criteria’ , which would be something like |xn+1-xn| < e, where e was an estimate of relative error.
How is this applied on ML/DL algos we’re learning? Or are we skipping on calculating a ‘stop criteria’ because we’re stipulating how many iterations or epochs to do?

Again, many thanks Jovian and FCC for such a well put course!

4 Likes

Hey @dan-motp check this out. I have tried to answer a similar question in this post. Lecture 1 - PyTorch Basics & Linear Regression

2 Likes

@dan-motp…the stop criterion here is the number of epochs u train the model for.

However, u can iteratively keep on training the model until, the loss estimate is below than your set expectation. But , that will be computationally expensive .

Best, is to plot a graph for trend of epochs against loss estimate…so that u can guess approx how many epochs will bring you to your desired loss estimate!

3 Likes

Hey @kamal You can use download zip button to get the notebook directly. image
Or do a clone of the notebook https://jovian.ml/docs/user-guide/reproduce.html#clone

3 Likes

I am trying to generate a data set of imaged using flow in Keras, but labels numpy dimensions are not matching and not getting generated do you know how to overcome this issue ?

1 Like

Yes, the attendance requirement has been removed.

4 Likes

How to know if I passed the first assignment?

3 Likes

So grateful to Aakash for this

1 Like

@aakashns I have a question. Will the certificate provided be a certificate of participation or something, or is it going to be an ACTUAL certificate, like the ones you get on Udemy, Coursera etc.?

5 Likes

The deadline has been extended. You’ll be receiving a mail regarding the acceptance/rejection of your assignment within next week.

2 Likes

If our assignment is rejected, will we have an option to retry the assignment?

4 Likes

Joining the 2nd lecture with the community!

3 Likes

Amazing stats from lecture1 !
Great stuff. Thanks

3 Likes