Deep Learning with PyTorch Live Course - Tensors, Gradient Descent & Linear Regression (Part 1 of 6)
video description
Date: 2022-03-14
Related videos
Comments and reviews: 10
dheeraj
-1:44.. the idea of having Floating point Weights is to find the Local minima of the Gradient Equation along with learning rate tuning as well , if the gradients is in integers the steps taken by the Gradient equation during epoch is large , may end up in loosing momentum to find global minima . to optimize the objective function along with learning rate , batch processing, momentum are needed to better perform the objective function i.e., Linear regression in turn results in good prediction model
reply
-1:44.. the idea of having Floating point Weights is to find the Local minima of the Gradient Equation along with learning rate tuning as well , if the gradients is in integers the steps taken by the Gradient equation during epoch is large , may end up in loosing momentum to find global minima . to optimize the objective function along with learning rate , batch processing, momentum are needed to better perform the objective function i.e., Linear regression in turn results in good prediction model
reply
Vinay
sir,
I came to know about this course just today,can i join from now?
as i already have experience in deep learning from coursera but i want some hand on experience with pytorch,
i already filled the google form and joined in the community in jovian what else i have to do in order to be a part of it?
reply
sir,
I came to know about this course just today,can i join from now?
as i already have experience in deep learning from coursera but i want some hand on experience with pytorch,
i already filled the google form and joined in the community in jovian what else i have to do in order to be a part of it?
reply
Muhammad
At 1:19:30, it is said that if the derivative is negative decreasing the weight will increase the loss but at 1:28:00 the weight is only decreased (w -= w.grad - 1e -5) despite the fact that the derivative is negative. I am confused here. Kindly clear this point.
reply
At 1:19:30, it is said that if the derivative is negative decreasing the weight will increase the loss but at 1:28:00 the weight is only decreased (w -= w.grad - 1e -5) despite the fact that the derivative is negative. I am confused here. Kindly clear this point.
reply
Sam
At 1:24:11, if I replace
w-= w.grad-1e-5
by
w = w - w.grad-1e-5
.. and similarly for updating b
I get an error then at w.grad.zero_()
AttributeError: 'NoneType' object has no attribute 'zero_'
Any hints or comments on this?
reply
At 1:24:11, if I replace
w-= w.grad-1e-5
by
w = w - w.grad-1e-5
.. and similarly for updating b
I get an error then at w.grad.zero_()
AttributeError: 'NoneType' object has no attribute 'zero_'
Any hints or comments on this?
reply
Jomiloju
Good day
I am interested in learning artificial intelligence I have already watched your Python Data science which of your video should I watch next on path of learning artificial intelligence.
Thank you while awaiting your reply.
reply
Good day
I am interested in learning artificial intelligence I have already watched your Python Data science which of your video should I watch next on path of learning artificial intelligence.
Thank you while awaiting your reply.
reply
Aman
I have just started following the course on YouTube. I am interested to pursue it further for a certification. Can I please continue with the assignments and lectures to be eligible for the certificate at the end of this course?
reply
I have just started following the course on YouTube. I am interested to pursue it further for a certification. Can I please continue with the assignments and lectures to be eligible for the certificate at the end of this course?
reply
Shaik
hey can anyone help me how i can use pytorch in jupyter?
i downloaded using pip in my command prompt but when i'm trying to import in jupyter it is showing no module found
reply
hey can anyone help me how i can use pytorch in jupyter?
i downloaded using pip in my command prompt but when i'm trying to import in jupyter it is showing no module found
reply
Zolt-n
Dear Sir! When will be the next certification course?, because we are ( lot of us) was late. For me will be really important, thank u in advance for your answer
reply
Dear Sir! When will be the next certification course?, because we are ( lot of us) was late. For me will be really important, thank u in advance for your answer
reply
Muhammad
how to plot that graph in which you explained us the gradient etc.
if this is something u r going to teach in future kindly let me know..
Thanks :)
reply
how to plot that graph in which you explained us the gradient etc.
if this is something u r going to teach in future kindly let me know..
Thanks :)
reply
Minakshee
You're actually a good teacher. feeling satisfied after learning from actual patient teacher and not just a random youtuber :-)
reply
You're actually a good teacher. feeling satisfied after learning from actual patient teacher and not just a random youtuber :-)
reply
Add a review, comment
Other channel videos
