Update - Lecture 2 Notebook of Deep Learning with Pytorch course

How to avoid HTTP-Error 503 while downloading the MNIST Dataset

If you are facing HTTP-Error 503 while downloading the MNIST Dataset it may be due to the website which is temporarily or permanently unavailable. To avoid this error you have to update the following code from the notebook.

Code to be Updated(where you are getting the HTTP Error):

# Download training dataset
dataset = MNIST(root='data/', download=True)

Updated code

!wget www.di.ens.fr/~lelarge/MNIST.tar.gz
!tar -zxvf MNIST.tar.gz
root_dir = './'
dataset = torchvision.datasets.MNIST(root=root_dir,download=True)

In addition to that you might face an error wherever the path of root argument is set as root='data/', whenever you see the root argument just make sure to change it to root=root_dir, or to the current working directory.

Anybody having any issues with matplotlib after importing pytorch ? I’m not running Aakash Bhaiya’s notebook, rather I’m trying to practice the same codes on a local notebook, that I created. But every time I call plt.imshow() to plot some image from MNIST dataset, the kernel is dying, and restarting. Literally stuck on this for hours now, I tried troubleshooting on google too, didn’t help much. Can anybody please help ? TIA.

Probably you are doing something wrong in plotting. Please chek the code for plt.