Competition for course "Event detection"
Init conda environment from .yml
file:
conda env create -f environment.yml
You need torchaudio for this project. To download additional dependencies follow these instructions:
https://github.com/pytorch/audio
This notebook requires NVidia GPU with CUDA kernetls. If do not have one, then this .yml
file does not suit you and you have to create environment by yourself. It it easy, just do:
conda create -n '<new_env>' python=3.7
-- create environment<new_env>
withpython3.7
onboardconda activate <new_env>
-- activate environmentpip install <package name>
-- install all required dependencies withpip
.
You can recognize dependencies looking at import
block. Also, correct the code in notebook, hiding all .cuda()
in it.