Decentralized compute layer
 
 
Go to file
Konstantine Tsafatinos 9f5a70ee11 address comments, add compute logic to support transformers 2023-06-13 09:17:28 -04:00
docker Add .ini example & new account config 2023-05-27 22:42:27 -03:00
skynet address comments, add compute logic to support transformers 2023-06-13 09:17:28 -04:00
tests Separate worker into submodules 2023-06-04 17:51:43 -03:00
.dockerignore Rework dgpu client to be single task 2022-12-17 11:39:42 -03:00
.gitignore Snappier dgpu, fix captioning & gitignores 2023-05-29 19:03:39 -03:00
LICENSE Update LICENSE 2023-04-09 19:50:08 -03:00
README.md Update README.md 2023-05-30 00:41:32 -03:00
build_docker.sh Use leap as network + auth layer, we decentarlized now 2023-05-22 06:10:51 -03:00
pytest.ini Use leap as network + auth layer, we decentarlized now 2023-05-22 06:10:51 -03:00
requirements.cuda.0.txt Separate cuda build in a third step 2023-01-16 00:26:04 -03:00
requirements.cuda.1.txt First decupled architecture, still working on integrating tractor gpu workers 2022-12-10 18:18:03 -03:00
requirements.cuda.2.txt Separate cuda build in a third step 2023-01-16 00:26:04 -03:00
requirements.txt Update py-leap to 1a14 2023-06-04 01:11:32 -03:00
setup.py Add img2img mode 2023-01-15 23:42:45 -03:00
skynet.ini.example Separate worker into submodules 2023-06-04 17:51:43 -03:00

README.md

skynet

decentralized compute platform

To launch a worker:

# create and edit config from template
cp skynet.ini.example skynet.ini

# create python virtual envoirment 3.10+
python3 -m venv venv

# enable envoirment
source venv/bin/activate

# install requirements
pip install -r requirements.txt
pip install -r requirements.cuda.0.txt
pip install -r requirements.cuda.1.txt
pip install -r requirements.cuda.2.txt

# install skynet
pip install -e .

# test you can run this command
skynet --help

# to launch worker
skynet run dgpu