Hi all I wanted to share my lesson 6 notes. As @timlee mentioned alot to cover in this lesson. I hope this will be of benefit to some. 
lesson6-rnn_notes.pdf (1.4 MB)
13 Likes
Very nice! Maybe add a link here from the wiki thread?
2 Likes
Done! 
It was Cool… Thanks @amritv for ur effort…
1 Like
If you don’t mind an ignorant question, how do you make these? Do you take notes during the live feed? Do you re-watch the video and make notes? Do you make your own notes like this? Is there a special/specific app you use for these type of notes?
@bhollan, I make the notes after re-watching the videos. Making these notes helps me better understand the code and I also use this process for other stuff. No special apps, I use acrobat pro, save the ipynb as an html and then use acrobat pro to save as a .pdf. Acrobat allows the use of the pen or finger to annotate the notes. Hope that helps.
1 Like
Yup! Thanks!
You might want to update your notes for the multi-output model loss function evaluator. At 2:03:20 Jeremy says that nh is 256. However, at 2:05:20 he says that nh is 84. I checked, and the correct size is 84 (which makes sense since you’re evaluating the loss on your prediction of 1 of 84 classes).
1 Like
@DavidBressler thanks for the info. Ill update this as soon as i can. 
1 Like
My notes if someone is interested to read same thing again a little bit differently.
Hey, I have a question on the part where we are creating input.
c1_dat = [idx[i] for i in range(0, len(idx)-1-cs, cs)]
c2_dat = [idx[i+1] for i in range(0, len(idx)-1-cs, cs)]
c3_dat = [idx[i+2] for i in range(0, len(idx)-1-cs, cs)]
c4_dat = [idx[i+3] for i in range(0, len(idx)-1-cs, cs)]
When we specify the range, why do we need to substract cs from the len(idx) ?