KISS Local Server (Pop!_OS)

Apologies for my english. I’m from Denmark and really only good at Danish :slight_smile:

I’m running Pop!_OS 20.04 (Ubuntu based distro) and everything is working without any complex configuration.

My hardware:
AMD Ryzen 5 2400G
NVIDIA RTX 2060 Super

PS. You can paste to the Terminal with Ctrl-Shift-V. The key “Super” is the Windows key.

1. Install Pop!_OS 20.04 LTS NVIDIA

https://pop-iso.sfo2.cdn.digitaloceanspaces.com/20.04/amd64/nvidia/13/pop-os_20.04_amd64_nvidia_13.iso

2. Open the Terminal

Press Super-T

3. Install Miniconda

wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
bash Miniconda3-latest-Linux-x86_64.sh
rm Miniconda3-latest-Linux-x86_64.sh

4. Close and reopen the Terminal

Press Super-Q
Press Super-T

5. Create you FastAI Conda environment

conda update conda
conda create --name fastai
conda activate fastai

6. Install all the fancy Deep Learning software and Jyputer Notebook

conda install -c pytorch pytorch==1.7.0 torchvision==0.8.1 cudatoolkit==10.2.89
conda install -c fastai fastai==2.1.2 fastcore==1.3.1 fastbook==0.0.11

7. Make a directory for the course files

mkdir Repos
cd Repos

8. Get the course files

git clone https://github.com/fastai/fastbook

9. Start Jupyter Notebook

jupyter notebook

10. Have fun and be smart in the company of Jeremy Howard

11. Future start of Jupyter Notebook

Press Super-T

conda activate fastai
cd Repos
jupyter notebook

2 Likes

A little update including TensorFlow 2.x

#open Teminal
sudo apt update
sudo apt upgrade
sudo apt install nvidia-container-runtime

wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
bash Miniconda3-latest-Linux-x86_64.sh
rm Miniconda3-latest-Linux-x86_64.sh

#exit Terminal
#open Terminal

conda update conda

conda create --name tensorflow
conda activate tensorflow
conda install tensorflow-gpu
conda install jupyter
conda deactivate

conda create --name fastai
conda activate fastai
conda install -c pytorch pytorch torchvision cudatoolkit
conda install -c fastai fastai fastcore fastbook
conda deactivate