How do I install PyTorch v1.0.0+ on Google Colab?

Question

PyTorch v1.0.0 stable was released on 8 December 2018 after being announced 7 months earlier.

I want get a version optimised for the hardware that my IPython kernel is running on.

How do I get this version on Google Colab?

score 21 · Accepted Answer · edited Jan 02 '19 at 08:36

21

try the following code snippet (it works equally for the runtime with or without gpu)

!pip install -q torch==1.0.0 torchvision

to check the version

import torch
print(torch.__version__)

here you have the version 1.0.0

UPDATE

!pip install torch

Works fine now, as the most stable version is 1.0.0

edited Jan 02 '19 at 08:36

Arunava

353
1
4
13

answered Dec 14 '18 at 10:01

ashunigion

1,769
1
11
15

1

Will this take into account the version of CUDA running on the system for optimisation? – Tom Hale Dec 17 '18 at 07:58
I think it does, it tried `torch.backends.cudnn.version()` and the output was `7401` and `torch.backends.cudnn.enabled == True` the output was `true`. I used the colab GPU runtime. – ashunigion Dec 17 '18 at 10:28
1

Great! I also get `7410` even though [my solution](https://stackoverflow.com/a/53775509/5353461) is based on Google's colab code snippet and has `cu90` in the generated download URL. – Tom Hale Dec 19 '18 at 05:10

score 5 · Answer 2 · answered Dec 14 '18 at 07:53

With version 1.0.0, PyTorch changed the download URL format from:

https://download.pytorch.org/whl/cu92/torch-1.0.0-cp36-cp36m-linux_x86_64.whl

to

https://download.pytorch.org/whl/cu90/torch-1.0.0-cp36-cp36m-linux_x86_64.whl

(The change is in the CUDA version part, where cu92 changes to cu90.)

To programmatically generate that URL, I used the following code:

from os.path import exists
from wheel.pep425tags import get_abbr_impl, get_impl_ver, get_abi_tag

platform = '{}{}-{}'.format(get_abbr_impl(), get_impl_ver(), get_abi_tag())
cuda_output = !ldconfig -p|grep cudart.so|sed -e 's/.*\.\([0-9]*\)\.\([0-9]*\)$/cu\10/'    
accelerator = cuda_output[0] if exists('/dev/nvidia0') else 'cpu'

torch_url=f"http://download.pytorch.org/whl/{accelerator}/torch-{version}-{platform}-linux_x86_64.whl"
version='1.0.0'

!pip install -U {torch_url} torchvision

You can then change the version variable as desired as newer versions are released.

score 3 · Answer 3 · answered Sep 04 '19 at 05:49

3

You can now just

import torch

No need for additional installation.

answered Sep 04 '19 at 05:49

Linh Le Vu

436
4
7

score 2 · Answer 4 · answered Jun 06 '19 at 03:01

2

For version 1.1.0, this works

!pip install -q torch==1.1.0 torchvision

answered Jun 06 '19 at 03:01

Oladimeji Mudele

63
1
1
5

score 2 · Answer 5 · answered Jul 16 '20 at 11:22

2

It worked for me you can try

!pip install torch
!pip install torchvision
!pip install mxnet-cu101

answered Jul 16 '20 at 11:22

Maheep

767
10
6

1

Actually, when I was installing pytorch using the 1st two commands it was showing error when I was importing but when I installed mxnet-cu101 it worked. – Maheep Jul 19 '20 at 15:27

score 1 · Answer 6 · answered May 13 '20 at 19:38

1

Here is a code to install pytorch. You can change it to whatever version you want.

!pip3 install http://download.pytorch.org/whl/cu92/torch-1.0.0-cp36-cp36m-linux_x86_64.whl

answered May 13 '20 at 19:38

Payam Jome Yazdian

368
3
8

How do I install PyTorch v1.0.0+ on Google Colab?

6 Answers6

Linked