PyTorch

Back to Index

Congratulations! You've reached the core step. 

Before typing anything, you can watch this video, which was where I found my way to make PyTorch work.

Now, let's type the commands to install the base for the Deep Learning libs:

• conda activate DeepFluid
• pip3 install ipykernel
• python3 -m ipykernel install --user --name DeepFluid --display-name "DeepFluid"

Next, the correct versions to work with our net:

• conda install -c pytorch pytorch==1.6.0 torchvision==0.7.0 cudatoolkit=10.1
• python3 -c "import torch; print(torch.cuda.is_available(), torch.cuda.current_device(), torch.cuda.device_count(), torch.cuda.get_device_name(0))"
• True 0 1 NVIDIA GeForce RTX 2060 SUPER

To help us visualize some data, type the following, but it is optional:

• conda install -c conda-forge tensorboard
• python3 -c "from torch.utils import tensorboard"
• Doesn't print anything, just works without errors.

Note: I like to type python3 when running commands, just to make sure it is version 3.

<- Back

Next ->