iampukar / deep-learning-in-computer-vision

Course work from Deep Learning in Computer Vision from National Research University Higher School of Economics
27 stars 9 forks source link

can you also provide the quiz answers i am really stuck. please help !!! #2

Open mansi-98 opened 4 years ago

tranvansanghust commented 4 years ago

Dear @mansi-98 , @gitosam Do you have free time. Because I have to complete the course by Sep, 20. So, Please help me to review my assignments: Week2: https://www.coursera.org/learn/deep-learning-in-computer-vision/peer/STsNv/facial-keypoints-notebook/review/jdm2Gfb_Eeq0VArkASuwDw Week3: https://www.coursera.org/learn/deep-learning-in-computer-vision/peer/8MWx2/face-detection-notebook/review/enmv1vb_EeqK6w7Z5Kuhgw Week4: https://www.coursera.org/learn/deep-learning-in-computer-vision/peer/dFypQ/face-recognition-notebook/review/lk4B__cVEeq9SA6hJWNb9Q Week5: https://www.coursera.org/learn/deep-learning-in-computer-vision/peer/9h2K8/gan-notebook/review/n92qDfcrEeq0VArkASuwDw

Thanks!

gsoumia commented 3 years ago

@gitosam did you make it with week 3 I am also stack in the quiz

mansi-98 commented 3 years ago

I tried to review your assignment but the link didn't work for me. Also i haven't completed the course yet.

On Thu 24 Dec, 2020, 9:51 PM gsoumia, notifications@github.com wrote:

@gitosam https://github.com/gitosam did you make it with week 3 I am also stack in the quiz

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/iampukar/deep-learning-in-computer-vision/issues/2#issuecomment-750921153, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJX72V6BWUX7TUM2SZECMADSWNTARANCNFSM4MTB25ZA .

gsoumia commented 3 years ago

@mansi-98: could you help me the quiz of week 3 question 5 or 2 please

mansi-98 commented 3 years ago

I dropped the course as i was stuck. All the answer that i was able to mark were with the help of discussion form. You may search in discussion form. Many people are stuck.

On Thu 24 Dec, 2020, 9:57 PM gsoumia, notifications@github.com wrote:

@mansi-98 https://github.com/mansi-98: could you help me the quiz of week 3 question 5 or 3 please

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/iampukar/deep-learning-in-computer-vision/issues/2#issuecomment-750922179, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJX72V6JLT6BFEH4HURNFYLSWNTW7ANCNFSM4MTB25ZA .

gsoumia commented 3 years ago

I only need one question to have 80% the pb is that I tried a lot of solutions and no thing work , I have extend the cours duration 3 times until now, thanks a lot.

gitosam commented 3 years ago

@gitosam did you make it with week 3 I am also stack in the quiz

yes @gsoumia

gsoumia commented 3 years ago

@gitosam could you help me with juste one question please? either that : Consider face detector have Miss rate 0.40 for FPPI = 10^-1. We are working with dataset that has 5 faces on each image in average. What Precision and Recall corresponds to this parameters? Enter precision and recall values in percentages with space:

or that one: We work with video of size 1024x768 pixels and 25 fps. We use sliding window object detector with window size 20x20 pixels and stride 2, 1 image scale (i.e. without pyramid). What should be false positive rate of the detector s.t. detector output false positive less frequently than 1 time per second? Round answer with 1e-07 precision.

gitosam commented 3 years ago

@gsoumia Consider face detector have Miss rate 0.40 for FPPI = 10^-1. We are working with dataset that has 5 faces on each image in average. What Precision and Recall corresponds to this parameters? Enter precision and recall values in percentages with space: Answer: 97 60

gsoumia commented 3 years ago

Thanks a lot I will try it actually I was sur about 90 60 could you explain the answer to me please? I know that: precision is 1 - FPPI and recall is 1 - miss rate

gitosam commented 3 years ago

@gsoumia FPPI = FP/ no. of images(N)=0.1 Ground truth = 5N TP = recall 5N = 0.6 5N = 3N Percision = TP/(TP + FP) = 3N/(3N+0.1N)=3/3.1N = 0.9677 i hope this will help..

gsoumia commented 3 years ago

thanks a lot

gitosam commented 3 years ago

You're welcome.