Torch install failure when using requirements.txt

[drewbyh]

Issue description

When running 'pip install -r requirements.txt' I get the following error on the Torch install:

ERROR: Could not find a version that satisfies the requirement torch<1.2.0,>=1.1.0 (from -r requirements.txt (line 21)) (from versions: 0.1.2, 0.1.2.post1, 0.1.2.post2) ERROR: No matching distribution found for torch<1.2.0,>=1.1.0 (from -r requirements.txt (line 21))

All installs before Torch are successful.

Code example/repro steps

ERROR: Could not find a version that satisfies the requirement torch<1.2.0,>=1.1.0 (from -r requirements.txt (line 21)) (from versions: 0.1.2, 0.1.2.post1, 0.1.2.post2) ERROR: No matching distribution found for torch<1.2.0,>=1.1.0 (from -r requirements.txt (line 21))

Expected behavior

Required Torch version should install correctly.

Screenshots

If applicable, add screenshots to help explain your problem. No screenshots of code!

System info

• How you installed Snorkel (conda, pip, source):
• Build command you used (if compiling from source): pip install -r requirements.txt
• OS: Windows 10 Build 1809
• Python version: 3.8
• Snorkel version:
• Versions of any other relevant libraries:

Additional context

Add any other context about the problem here.

[stenpiren]

I have exactly the same problem. Not just when installing from requirements.txt but a simply pip install snorkel falls with the same error code:

Could not find a version that satisfies the requirement torch<1.2.0,>=1.1.0 (from snorkel) (from versions: 0.1.2, 0.1.2.post1, 0.1.2.post2) No matching distribution found for torch<1.2.0,>=1.1.0 (from snorkel)

Also, the requirements.txt fail in most of the tutorials. Same error code.

[MortenKP]

I have the same problem as well.

[MortenKP]

Installing one of these before "pip install snorkel" solved it for me:

CUDA 9.0

conda install pytorch==1.1.0 torchvision==0.3.0 cudatoolkit=9.0 -c pytorch

CUDA 10.0

conda install pytorch==1.1.0 torchvision==0.3.0 cudatoolkit=10.0 -c pytorch

CPU Only

conda install pytorch-cpu==1.1.0 torchvision-cpu==0.3.0 cpuonly -c pytorch

Source: https://pytorch.org/get-started/previous-versions/

[henryre]

Thanks for posting your solution @MortenKP ! We have a note on using conda with PyTorch in the README but feel free to propose a clearer description! Going to close this out, but feel free to re-open if needed!