That could very well be the case - since the first notebook had the many bear downloads and such. My second notebook doesn’t have that baggage as I moved on with the .pkl file only, so, this could very well be the case.
I just created a Gradient notebook, was running into allotment issues with Colab. However, there is no /datasets or /storage folder on creation anymore. Can I just create one manually?
So they’ve recently upgraded the V1 instances to V2 instances. The /storage folder (and my advice above) pertains to V1 instances.
I don’t think you need a /storage folder anymore. But please confirm with Paperspace’s customer support (just to be on the safe side).
Paperspace has a habit of losing customer data. If you don’t want to lose your notebooks & fine-tuned models, then download your data to your laptop after every session. Even better, avoid using Paperspace altogether.
1 Like
Were you able to resolve this? I am still stuck at this basic step.
How do you get the folder “fastbook”. All I see is ipynb files for the chapters. Do you have a clean set of instructions for this?
Hi folks,
I am a bit lost. I am following the instructions for paperspace as mentioned in Paperspace Gradient | Practical Deep Learning for Coders but when I select the fastai runtime from gradient, I dont’t see any course material for v4. All I see is the ipynb for the course book. See attached image.
Can some one please help me get started? I just want to know first few steps
1 Like
This is the V4 course. It follows the book. Check clean folder for notebooks with code only. Happy learning!
Can you address the errors? I’m having the same issue too and instructions to upgrade pip give me a syntax error
What error do you get?
using
python -m pip install --upgrade pip
pip install -Uqq fastbook --use-feature=2020-resolver
i get:
File "<ipython-input-20-ce7bdc82cd20>", line 2
python -m pip install --upgrade pip
^
SyntaxError: invalid syntax
if you want to run these commands from notebook you need to type ! in front:
! python -m pip install --upgrade pip
! pip install -Uqq fastbook --use-feature=2020-resolver
1 Like
thanks, that worked
Hi I’ve seen that some user are having the same problem I am facing currently with Paperspace, and I tried your solution but still I’ve got the same problem. Here is a screenshot of my issue :
Don’t see the full error message. But paperspace have issue at the moment with torch / torch vision install. You can try installing latest versions of these libs and rasing ticket at paperspace so they can fix the error.
Hi @dkobran !
As I asked here, I’d love to use Paperspace Gradient’s GPU, while coding in my local VS Code. Kind of the opposite to what you showed in the message I replied to 
Does this make any sense to you, is it possible to do something like that? I’m not a fan of coding in the browser (you can find more details in the longer post I linked to)
Thanks a lot in advance!
edit: solution given by JH 
→ Visual Studio Code Remote Development
Hi @maraoz,
Just to add to the solution, you can connect a python notebook in your local VSCode instance to the Paperspace Gradient Notebook Jupyter kernel. Documentation on how to do that can be found here.
Let me know if you have any further questions about this setup.
2 Likes
2.5 years later, I came across this issue at 5am was very frustrated, but I found your solution and it worked. Thank you!
Hi how do we show the files on the Gradient server in VS Code? I can connect and create a new notebook. When I use !ls I can see my files, but they dont appear in my VS Code Explorer, so I can’t open them.
Any tips?
Currently the connection instructions and token we have in the Gradient Notebook only allows you to connect to the remote Jupyter kernel which will allow you to run a python notebook in VSCode but doesn’t give you full access like ssh using VSCode Remote Explorer would.
However, VSCode just released the following instructions which does provide a way to tunnel in to your Gradient Notebook from your local VSCode instance. This should give you the experience you are looking for (e.g. seeing system files in VSCode and being able to open/edit them). Making Remote Development Even Better
Let me know if you have any questions about the above